JP2014082391A - Semiconductor device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a highly reliable semiconductor device using an oxide semiconductor by inhibiting variation in electrical characteristics of the semiconductor device.SOLUTION: In a semiconductor device, by supplying oxygen to a base insulation layer arranged under an oxide lamination including an oxide semiconductor layer and a gate insulation layer arranged above the oxide lamination to a channel formation region, possible oxygen deficiency to be formed in a channel region is compensated. A cross section of the oxide lamination is formed into a shape including a curved surface. In addition, by inhibiting oxygen abstraction from the oxide semiconductor layer by a source electrode layer or a drain electrode layer near the channel region formed in the oxide semiconductor layer, oxygen deficiency in the channel region is inhibited. Further, by forming a protective insulation layer functioning as a barrier layer which has a low hydrogen content and a low oxygen permeability on the gate electrode layer, oxygen desorption from the gate insulation layer and/or the base insulation layer is inhibited thereby to effectively supply oxygen to the channel formation region.

Description

The invention disclosed in this specification and the like relates to a semiconductor device and a method for manufacturing the semiconductor device.

Note that in this specification and the like, a semiconductor device refers to all devices that can function by utilizing semiconductor characteristics, and an electro-optical device, an image display device, a semiconductor circuit, and an electronic device are all semiconductor devices.

A technique for forming a transistor using a semiconductor thin film formed over a substrate having an insulating surface has attracted attention. The transistor is widely applied to electronic devices such as an integrated circuit (IC) and an image display device (also simply referred to as a display device). A silicon-based semiconductor material is widely known as a semiconductor thin film applicable to a transistor, but an oxide semiconductor has attracted attention as another material.

For example, a technique for manufacturing a transistor using zinc oxide or an In—Ga—Zn-based oxide semiconductor as an oxide semiconductor is disclosed (see Patent Documents 1 and 2).

JP 2007-123861 A JP 2007-96055 A

An object of one embodiment of the present invention is to impart favorable electrical characteristics to a semiconductor device including an oxide semiconductor.

Another object of one embodiment of the present invention is to provide a highly reliable semiconductor device in which variation in electrical characteristics of a semiconductor device including an oxide semiconductor is suppressed.

In the case where a transistor is formed using an oxide semiconductor, oxygen vacancies are given as one of supply sources of carriers of the oxide semiconductor. If there are many oxygen vacancies in the oxide semiconductor including the channel formation region of the transistor, electrons are generated in the channel formation region, causing the transistor to be normally on, an increase in leakage current, and the threshold voltage due to stress application. It becomes a factor which causes defective electrical characteristics such as shift. In the oxide semiconductor layer, hydrogen, silicon, nitrogen, carbon, and metal elements other than main components are impurities. For example, hydrogen forms a donor level in the oxide semiconductor layer and increases the carrier density. In addition, silicon forms impurity levels in the oxide semiconductor layer, and the impurity levels may serve as a trap, which may deteriorate the electrical characteristics of the transistor.

Therefore, in order to obtain stable electrical characteristics in a semiconductor device including an oxide semiconductor, measures must be taken to reduce oxygen vacancies in the oxide semiconductor layer and reduce the concentration of impurities such as hydrogen and silicon. Is required.

Therefore, in the semiconductor device of one embodiment of the present invention, oxygen is supplied from the base insulating layer provided below the oxide semiconductor layer and the gate insulating layer provided above the oxide semiconductor layer to the channel formation region. Thus, oxygen vacancies that can be formed in the channel region are compensated. In addition, oxygen vacancies in the channel region are suppressed by suppressing extraction of oxygen from the oxide semiconductor layer by the source or drain electrode layer in the vicinity of the channel region formed of the oxide semiconductor layer. Furthermore, desorption of oxygen from the gate insulating layer and / or the base insulating layer is performed by forming a protective insulating layer that functions as a barrier layer with a low hydrogen content and low oxygen permeability over the gate electrode layer. And oxygen is effectively supplied to the channel formation region.

In addition, an oxide layer including one or more metal elements included in the oxide semiconductor layer is provided in contact with the upper and lower sides of the oxide semiconductor layer forming the channel. Thus, the channel can be separated from the gate insulating layer. In addition, since an interface state is hardly formed at the interface between the oxide layer and the oxide semiconductor layer, variation in electrical characteristics such as a threshold voltage of the transistor can be reduced.

Further, a cross section of the oxide stack including the oxide semiconductor layer and the stack of the oxide layers provided in contact with the upper and lower sides of the oxide semiconductor layer has a curved surface. By having a curved surface in the cross section of the oxide stack, the coverage of a film formed over the oxide stack can be improved. Thus, a film formed over the oxide stack can be formed uniformly, and an impurity element enters the oxide stack from a region with a low film density or a region where no film is formed, thereby improving the characteristics of the semiconductor device. Deterioration can be suppressed and a semiconductor device with stable characteristics can be obtained. Note that it is particularly preferable that the lower end portion of the oxide stack or the lower end portion and the upper end portion have curved surfaces.

The semiconductor device according to one embodiment of the present invention has the above-described structure, so that the impurity concentration of the oxide semiconductor layer functioning as a channel (which serves as a main path of carriers) is reduced and highly purified intrinsic. Can do. High purity intrinsic refers to making an oxide semiconductor layer intrinsic or substantially intrinsic. Note that in this specification and the like, the carrier density of the oxide semiconductor layer is less than 1 × 10 ¹⁷ / cm ^3, less than 1 × 10 ¹⁵ / cm ³ , or less than 1 × 10 ¹³ / cm ³ when it is substantially intrinsic. It is. By making the oxide semiconductor layer highly purified and intrinsic, stable electrical characteristics can be imparted to the transistor.

More specifically, for example, the following configuration can be adopted.

One embodiment of the present invention includes a base insulating layer containing oxygen, an island-shaped oxide stack provided over the base insulating layer, a part of the top surface of the island-shaped oxide stack, and a side surface in the channel formation direction. A metal nitride film which is provided on each of the first source electrode layer and the first drain electrode layer which are in contact with each other, and which is provided on each of the first source electrode layer and the first drain electrode layer and which is in contact with part of the upper surface of the oxide stack. A second source electrode layer and a second drain electrode layer, and a second source electrode layer and a second drain electrode layer provided between the second source electrode layer and the second drain electrode layer. A gate insulating layer in contact with the upper surface of the oxide stack, a gate electrode layer overlapping with the oxide stack with the gate insulating layer interposed therebetween, and a protective insulating layer provided in contact with the gate insulating layer and the gate electrode layer. The oxide stack has at least a channel An oxide semiconductor layer, a first oxide layer provided between the oxide semiconductor layer and the base insulating layer, and a second oxide provided between the oxide semiconductor layer and the gate insulating layer And a third oxide layer provided in contact with side surfaces of the first oxide layer, the oxide semiconductor layer, and the second oxide layer, and the third oxide layer has a curved surface. The base insulating layer and the gate insulating layer have a region in contact with the outer peripheral portion of the island-shaped oxide stack, and the protective insulating layer is a semiconductor device having lower permeability to oxygen than the gate insulating layer.

One embodiment of the present invention includes an oxygen-containing base insulating layer, an island-shaped oxide stack provided over the base insulating layer, a part of the top surface of the island-shaped oxide stack, and a channel formation direction. Metal nitride that is provided on each of the first source electrode layer and the first drain electrode layer in contact with the side surface, and on the first source electrode layer and the first drain electrode layer, and in contact with part of the upper surface of the oxide stack. A second source electrode layer and a second drain electrode layer made of a film; and a second source electrode layer and a second drain electrode layer provided on the second source electrode layer and the second drain electrode layer. A gate insulating layer in contact with the upper surface of the oxide stack, and a gate electrode layer overlapping with part of the oxide stack, the second source electrode layer, and the second drain electrode layer through the gate insulating layer, The oxide stack has at least an acid that forms a channel. An oxide semiconductor layer, a first oxide layer provided between the oxide semiconductor layer and the base insulating layer, a second oxide layer provided between the oxide semiconductor layer and the gate insulating layer, The first oxide layer, the oxide semiconductor layer, and the third oxide layer provided in contact with the side surfaces of the second oxide layer, and the third oxide layer has a curved surface The base insulating layer and the gate insulating layer have a region in contact with the outer peripheral portion of the island-shaped oxide stack, and the protective insulating layer is a semiconductor device having lower oxygen permeability than the gate insulating layer.

In the above semiconductor device, the oxide semiconductor layer, the first oxide layer, and the second oxide layer include an In-M-Zn oxide (M is Al, Ti, Ga, Y, Zr, La, Ce, Nd or Hf), and the first oxide layer and the second oxide layer preferably have a larger atomic ratio of M to In than the oxide semiconductor layer.

In the above semiconductor device, the oxide semiconductor layer preferably includes a crystal part, and the c-axis of the crystal part is preferably parallel to the normal vector of the surface of the oxide semiconductor layer.

In the above semiconductor device, the first source electrode layer and the first drain electrode layer are made of a material that is more easily bonded to oxygen than the second source electrode layer and the second drain electrode layer. .

In the above semiconductor device, the concentration of hydrogen contained in the protective insulating layer is preferably less than 5 × 10 ¹⁹ cm ⁻³ .

According to one embodiment of the present invention, favorable electrical characteristics can be imparted to a semiconductor device including an oxide semiconductor.

According to one embodiment of the present invention, a change in electrical characteristics of a semiconductor device including an oxide semiconductor can be suppressed, and a highly reliable semiconductor device can be provided.

4A and 4B are a plan view and cross-sectional views illustrating one embodiment of a semiconductor device. 4A and 4B are a plan view and cross-sectional views illustrating one embodiment of a semiconductor device. FIG. 10 is a cross-sectional view illustrating one embodiment of a semiconductor device. 10 is a cross-sectional view illustrating an example of a method for manufacturing a semiconductor device. FIG. 10 is a cross-sectional view illustrating an example of a method for manufacturing a semiconductor device. FIG. FIG. 6 is a cross-sectional view of an oxide stack according to one embodiment of the present invention. FIG. 6 is a cross-sectional view of an oxide stack according to one embodiment of the present invention. 10 is a cross-sectional view illustrating a method for manufacturing an oxide stack according to one embodiment of the present invention. 10A and 10B are cross-sectional views illustrating a method for manufacturing an oxide stack according to one embodiment of the present invention. 10A and 10B are cross-sectional views illustrating a method for manufacturing an oxide stack according to one embodiment of the present invention. The figure which shows a SIMS measurement result. The figure which shows a SIMS measurement result. The figure which shows a SIMS measurement result. The figure which shows a SIMS measurement result. The figure which shows the measurement result of sheet resistance. The figure which shows the measurement result of sheet resistance. The figure which shows the measurement result of sheet resistance. The figure which shows a TDS analysis result. 10A and 10B are cross-sectional views illustrating a method for manufacturing an oxide stack according to one embodiment of the present invention. 10A and 10B are cross-sectional views illustrating a method for manufacturing an oxide stack according to one embodiment of the present invention. 6A and 6B illustrate a band structure of an oxide stack according to one embodiment of the present invention. 6A and 6B illustrate a band structure of an oxide stack according to one embodiment of the present invention. FIG. 10 is a circuit diagram of a semiconductor device of one embodiment of the present invention. 4A and 4B are a circuit diagram and a conceptual diagram of a semiconductor device of one embodiment of the present invention. FIG. 10 is a block diagram of a semiconductor device of one embodiment of the present invention. FIG. 10 is a block diagram of a semiconductor device of one embodiment of the present invention. FIG. 10 is a block diagram of a semiconductor device of one embodiment of the present invention. An electronic device to which the semiconductor device of one embodiment of the present invention can be applied.

Hereinafter, embodiments of the disclosed invention will be described in detail with reference to the drawings. However, the invention disclosed in this specification is not limited to the following description, and it is easily understood by those skilled in the art that modes and details can be variously changed. Therefore, the invention disclosed in this specification is not construed as being limited to the description of the embodiments below.

Note that in the structures of one embodiment of the present invention described below, the same portions or portions having similar functions are denoted by the same reference numerals in different drawings, and description thereof is not repeated. In addition, in the case where the same function is indicated, the hatch pattern is the same, and there is a case where no reference numeral is given.

In the present specification and the like, ordinal numbers such as “first” and “second” are used for avoiding confusion between components, and are not limited numerically.

Note that the functions of the “source” and “drain” of the transistor may be interchanged when a transistor with a different polarity is used or when the direction of current changes during circuit operation. Therefore, in this specification, the terms “source” and “drain” can be used interchangeably.

(Embodiment 1)
In this embodiment, an oxide stack including an oxide semiconductor layer included in the semiconductor device of one embodiment of the present invention will be described with reference to FIGS.

<Configuration of oxide stack>
6A to 6D illustrate a cross-sectional structure of the oxide stack 404. FIG. The oxide stack 404 includes a first oxide layer 404a, an oxide semiconductor layer 404b provided over the first oxide layer 404a, and a second oxide layer provided over the oxide semiconductor layer 404b. 404c and a third oxide layer 404d provided in contact with side surfaces of the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c. Note that the third oxide layer 404d has a curved surface.

In the oxide stack 404 illustrated in FIGS. 6A and 6B, the third oxide layer 404 d has one curved surface defined by an arbitrary curvature radius. In the oxide stack 404 illustrated in FIGS. 6C and 6D, the third oxide layer 404d has two or more curved surfaces defined by an arbitrary curvature radius.

As illustrated in FIGS. 6A and 6C, the oxide stack 404 includes the side surfaces of the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c and the first oxidation layer. The angle formed with the lower surface of the physical layer 404a may be substantially vertical, and may have an inclination (taper angle) as shown in FIGS. 6B and 6D.

Subsequently, an oxide layer constituting the end portion of the oxide stack will be described. Hereinafter, a curved surface included in the third oxide layer 404d will be described with reference to FIGS.

FIG. 7A is a cross-sectional view of the third oxide layer 404d corresponding to FIGS. 6A and 6B. The third oxide layer 404d illustrated in FIG. 7A has one curved surface with a radius of curvature defined by r. Note that the radius of curvature is equal to the radius of a curved contact circle (also referred to as a curvature circle).

FIG. 7B is a cross-sectional view of the third oxide layer 404d corresponding to FIGS. 6C and 6D. The third oxide layer 404d illustrated in FIG. 7B has two curved surfaces with a curvature radius defined by r. In addition, it is good also as a structure which has two or more curved surfaces from which a curvature radius differs.

The third oxide layer 404d illustrated in FIG. 7C has one curved surface with a radius of curvature defined by r. Note that the third oxide layer 404d may have two or more curved surfaces. In that case, the curvature radius of each curved surface may be the same or different.

<Oxide stack formation mechanism 1>
A formation mechanism of the oxide stack 404 including the third oxide layer 404d having a curved surface is described. Hereinafter, an example of a formation mechanism of the oxide stack 404 including the curved third oxide layer 404d will be described with reference to FIGS.

First, the first oxide film 436a provided over the base insulating layer 402, the oxide semiconductor film 436b provided over the first oxide film 436a, and the first oxide film 436b provided over the oxide semiconductor film 436b. An oxide stack including two oxide films 436c is prepared (see FIG. 8A).

Next, a resist mask 140 is formed over part of the second oxide film 436c (see FIG. 8B).

Next, the second oxide film 436c and the oxide semiconductor film 436b in a region where the resist mask 140 is not provided are etched by dry etching to expose the first oxide film 436a (FIG. 8C )reference).

Next, the exposed first oxide film 436a is etched by a dry etching method (see FIG. 9A). At this time, a reaction product of the first oxide film 436a is reattached to the side surface of the oxide stack, so that an oxide layer 436d which is a sidewall protective film (also called a rabbit ear) is formed. Note that the reaction product of the first oxide film 436a is reattached by a sputtering phenomenon and is reattached through plasma during dry etching.

Subsequently, the first oxide film 404a and the oxide layer 437d are formed by etching the first oxide film 436a. At this time, part of the base insulating layer 402 is also etched, so that the thickness of a region exposed from the oxide layer 437d and the resist mask 140 is reduced (see FIG. 9B).

Note that since the oxide layer 437d is a reaction product of the first oxide film 436a, components derived from the etching gas used during etching (such as chlorine and boron) remain. When the component reacts with moisture or the like in the air, the oxide layer 437d is further etched to be an oxide layer 438d (see FIG. 9C).

Next, a component derived from an etching gas remaining in the oxide layer 438d is removed by ashing treatment, whereby the third oxide layer 404d is formed (see FIG. 10A). The third oxide layer 404d subjected to the ashing treatment can have a lower carrier density than the oxide layer 438d.

Next, the resist mask 140 is removed (see FIG. 10B).

As described above, the oxide stack 404 including the third oxide layer 404d having a curved surface can be formed. Therefore, it can be seen that a dedicated photomask or the like is not necessary to form the curved third oxide layer 404d.

Here, the heat treatment is preferably performed in an atmosphere containing an oxidizing gas. Note that the oxidizing gas refers to a gas such as oxygen, nitrous oxide, or ozone. Through the heat treatment, oxygen vacancies in the first oxide layer 404a, the oxide semiconductor layer 404b, the second oxide layer 404c, and the third oxide layer 404d can be reduced. In particular, the carrier density of the third oxide layer 404d can be extremely reduced by the heat treatment.

Note that after the oxide stack 404 is formed, the base insulating layer 402 having a plurality of steps (here, two steps) may be formed by etching the base insulating layer 402 using the oxide stack 404 as a mask ( (See FIG. 10C).

<Oxide stack formation mechanism 2>
Hereinafter, an example different from the above of the formation mechanism of the oxide stack 404 including the curved third oxide layer 404d will be described with reference to FIGS.

First, the first oxide film 436a provided over the base insulating layer 402, the oxide semiconductor film 436b provided over the first oxide film 436a, and the first oxide film 436b provided over the oxide semiconductor film 436b. An oxide stack including two oxide films 436c is prepared (see FIG. 19A).

Next, a resist mask 140 is formed over part of the second oxide film 436c (see FIG. 19B).

Next, the second oxide film 436c, the oxide semiconductor film 436b, and the first oxide film 436a in a region where the resist mask 140 is not provided are etched by dry etching, so that the oxide layer 456c and the oxide film 436c are oxidized, respectively. A physical semiconductor layer 456b and an oxide layer 456a are used. At this time, part of the base insulating layer 402 is also etched, so that the thickness of a region exposed from the oxide layer 456a is reduced (see FIG. 19C).

Next, the oxide layer 456c, the oxide semiconductor layer 456b, and the oxide layer 456a are etched by a dry etching method, and the second oxide layer 404c, the oxide semiconductor layer 404b, and the first oxide layer 404a are respectively etched. Form. At this time, the reaction product of the oxide layer 456a is reattached to the side surface of the oxide stack to form an oxide layer 456d which is a sidewall protective film (also called a rabbit ear). Note that the reaction product of the oxide layer 456a is reattached by a sputtering phenomenon and is reattached via plasma during dry etching. At this time, part of the base insulating layer 402 is also etched (see FIG. 20A).

Note that since the oxide layer 456d is a reaction product of the oxide layer 456a, components derived from the etching gas used during etching (such as chlorine and boron) remain.

Next, the third oxide layer 404d is formed by removing the components derived from the etching gas remaining in the oxide layer 456d by an ashing process. The third oxide layer 404d subjected to the ashing treatment can have a lower carrier density than the oxide layer 456d.

After that, the resist mask 140 is removed (see FIG. 20B).

As described above, the oxide stack 404 including the third oxide layer 404d having a curved surface can be formed. Therefore, it can be seen that a dedicated photomask or the like is not necessary to form the curved third oxide layer 404d.

Here, the heat treatment is preferably performed in an atmosphere containing an oxidizing gas. Through the heat treatment, oxygen vacancies in the first oxide layer 404a, the oxide semiconductor layer 404b, the second oxide layer 404c, and the third oxide layer 404d can be reduced. In particular, the carrier density of the third oxide layer 404d can be extremely reduced by the heat treatment.

As described above, the third oxide layer 404d is formed from a reaction product of the first oxide film 436a that becomes the first oxide layer 404a. Therefore, the first oxide layer 404a and the third oxide layer 404d may not be distinguished by analysis or the like. In other words, the third oxide layer 404d may be an oxide layer having physical properties similar to those of the first oxide layer 404a. Therefore, the description of the first oxide layer 404a can be referred to when the physical properties of the third oxide layer 404d are not particularly described. In addition, the second oxide layer 404c may be indistinguishable from the third oxide layer 404d in the case where the structure is similar to that of the first oxide layer 404a.

The oxide stack 404 has a structure in which the oxide semiconductor layer 404b is covered with the first oxide layer 404a, the second oxide layer 404c, and the third oxide layer 404d. Accordingly, entry of impurities into the oxide semiconductor layer 404b can be reduced. In addition, since the oxide semiconductor layer 404b does not have a level with another oxide layer, carrier mobility can be increased.

<Configuration example of oxide stack>

The first oxide layer 404a and the second oxide layer 404c are oxide layers containing one or more metal elements included in the oxide semiconductor layer 404b.

In addition, the first oxide layer 404a and the second oxide layer 404c include one or more metal elements included in the oxide semiconductor layer 404b, and the energy at the lower end of the conduction band is less than that of the oxide semiconductor layer 404b. It is formed of an oxide semiconductor close to a vacuum level in a range of any of 05 eV, 0.07 eV, 0.1 eV, and 0.15 eV and any of 2 eV, 1 eV, 0.5 eV, and 0.4 eV. It is preferable.

The oxide semiconductor layer 404b is a layer represented by an In-M-Zn oxide containing at least indium, zinc, and M (metal such as Al, Ga, Ge, Y, Zr, Sn, La, Ce, or Hf). including. It is preferable that the oxide semiconductor layer 404b contain indium because carrier mobility of the transistor is increased.

The first oxide layer 404a under the oxide semiconductor layer 404b is represented by In-M-Zn oxide (metal such as Al, Ti, Ga, Ge, Y, Zr, Sn, La, Ce, or Hf). And an oxide layer having a higher atomic ratio of M than the oxide semiconductor layer 404b. Specifically, as the first oxide layer 404a, an oxide containing the above element at an atomic ratio higher than that of the oxide semiconductor layer 404b by 1.5 times or more, preferably 2 times or more, more preferably 3 times or more. Use physical layers. The above element is more strongly bonded to oxygen than indium, and thus has a function of suppressing generation of oxygen vacancies in the oxide layer. In other words, the first oxide layer 404a is an oxide layer in which oxygen vacancies are less likely to occur than in the oxide semiconductor layer 404b.

Further, as the second oxide layer 404c over the oxide semiconductor layer 404b, an In-M-Zn oxide (Al, Ti, Ga, Ge, Y, Zr, A metal such as Sn, La, Ce, or Hf), which includes an oxide layer having a higher atomic ratio of M than the oxide semiconductor layer 404b. Specifically, as the second oxide layer 404c, an oxide containing the above element at an atomic ratio higher than that of the oxide semiconductor layer 404b by 1.5 times or more, preferably 2 times or more, more preferably 3 times or more. Use physical layers.

That is, the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c each include at least indium, zinc, and M (Al, Ti, Ga, Ge, Y, Zr, Sn, La, Ce, or In the case of an In-M-Zn oxide containing a metal such as Hf, the first oxide layer 404a is formed of In: M: Zn = x ₁ : y ₁ : z ₁ [atomic ratio], an oxide semiconductor layer 404b represents In: M: Zn = x ₂ : y ₂ : z ₂ [atomic number ratio], and the second oxide layer 404c represents In: M: Zn = x _3: y _3: z ₃ [atomic number ratio]. _Then, it is preferable that y 1 / _{x 1} and _y 3 / _{x 3} is greater than _y 2 / _{x 2.} y ₁ / x ₁ and y ₃ / x ₃ are 1.5 times or more, preferably 2 times or more, more preferably 3 times or more than y ₂ / x ₂ . In this case, in the oxide semiconductor layer 404b, the y ₂ is at x ₂ or more electrical characteristics of the transistor can be stabilized. However, when y ₂ is 3 times or more of x ₂ , the field-effect mobility of the transistor is lowered. Therefore, y ₂ is preferably the same as x ₂ or less than 3 times.

Note that when the first oxide layer 404a is an In-M-Zn oxide, the atomic ratio of In to M is preferably less than 50 atomic% for In, more than 50 atomic% for M, and more preferably 25 atomic% for In. And M is 75 atomic% or more. In the case where the oxide semiconductor layer 404b is an In-M-Zn oxide, the atomic ratio of In to M is preferably that In is 25 atomic% or more, M is less than 75 atomic%, more preferably In is 34 atomic% or more, M is less than 66 atomic%. In the case where the second oxide layer 404c is an In-M-Zn oxide, the atomic ratio of In to M is preferably less than 50 atomic% for In, more than 50 atomic% for M, and more preferably 25 atomic% for In. And M is 75 atomic% or more. When the third oxide layer 404d is an In-M-Zn oxide, the atomic ratio of In to M is preferably less than 50 atomic% for In, more than 50 atomic% for M, and more preferably 25 atomic% for In. And M is 75 atomic% or more.

Note that the first oxide layer 404a and the second oxide layer 404c may be layers containing different constituent elements, or contain the same constituent elements in the same atomic ratio or in different atomic ratios. It is good also as a layer.

By providing the oxide layer which is less likely to cause oxygen vacancies than the oxide semiconductor layer 404b in contact with the upper and lower sides of the oxide semiconductor layer 404b, oxygen vacancies that can be generated in the oxide semiconductor layer 404b can be reduced. . The third oxide layer 404d in contact with the side surface of the oxide semiconductor layer 404b has a structure similar to that of the first oxide layer 404a or the second oxide layer 404c, and thus the side surface of the oxide semiconductor layer 404b. Oxygen vacancies can also be suppressed from occurring.

Oxygen vacancies included in the oxide semiconductor layer are manifested as localized levels that exist at deep energy positions in the energy gap of the oxide semiconductor. When carriers are trapped in such localized states, the reliability of the transistor is reduced, so that oxygen vacancies in the oxide semiconductor layer need to be reduced. In the oxide stack 404, oxygen vacancies in the oxide semiconductor layer 404b are reduced by providing oxide layers that are less likely to cause oxygen vacancies than the oxide semiconductor layer 404b in contact with the upper and lower sides of the oxide semiconductor layer 404b. be able to. For example, the oxide semiconductor layer 404b has an absorption coefficient of less than 1 × 10 ⁻³ / cm, preferably 1 × 10 ⁻⁴ / cm as measured by a constant current measurement method (CPM: Constant Photocurrent Method). Less than.

Note that the first oxide layer 404a and the second oxide layer 404c may be layers containing different constituent elements, or contain the same constituent elements in the same atomic ratio or in different atomic ratios. It is good also as a layer.

For the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c, for example, an oxide semiconductor containing indium, zinc, and gallium can be used.

The thickness of the first oxide layer 404a is 3 nm to 100 nm, preferably 3 nm to 50 nm. The thickness of the oxide semiconductor layer 404b is 3 nm to 200 nm, preferably 3 nm to 100 nm, more preferably 3 nm to 50 nm.

In the case where an insulating layer containing silicon is used as the base insulating layer 402 or the gate insulating layer 410 in contact with the oxide stack 404, silicon in the insulating layer or carbon that can be mixed into the insulating layer is used as the first oxide. The layer 404a or the second oxide layer 404c may be mixed up to several nanometers from the interface. When an impurity such as silicon or carbon enters the oxide semiconductor layer, an impurity level is formed, and the impurity level serves as a donor to generate electrons, so that the n-type may be obtained. However, if the thicknesses of the first oxide layer 404a and the second oxide layer 404c are larger than several nm, impurities such as mixed silicon and carbon do not reach the oxide semiconductor layer 404b. The effect of impurity levels is reduced.

Here, the concentration of silicon contained in the oxide semiconductor layer is 3 × 10 ¹⁸ / cm ³ or less, preferably 3 × 10 ¹⁷ / cm ³ or less. The concentration of carbon contained in the oxide semiconductor layer is 3 × 10 ¹⁸ / cm ³ or less, preferably 3 × 10 ¹⁷ / cm ³ or less. In the oxide stack 404 described in this embodiment, the first oxide layer 404a, the second oxide layer 404c, and the third oxide layer 404d surround the oxide semiconductor layer 404b serving as a carrier path. This is preferable because contamination into silicon or carbon which is a Group 14 element can be reduced in the oxide semiconductor layer 404b.

Note that the impurity concentration in the oxide semiconductor layer can be measured by a secondary ion analysis method (SIMS: Secondary Ion Mass Spectrometry).

<Band structure of oxide stack>
Hereinafter, the first oxide layer 404a, the oxide semiconductor layer 404b, the second oxide layer 404c, and the third oxide layer 404d included in the oxide stack 404 are formed using the band structure of the oxide stack 404. Will be described.

When the energy at the lower end of the conduction band of the first oxide layer 404a, the oxide semiconductor layer 404b, the second oxide layer 404c, and the third oxide layer 404d is EcS1, EcS2, EcS3, and EcS4, respectively, The first oxide layer 404a, the oxide semiconductor layer 404b, the second oxide layer 404c, and the third oxide layer 404d are selected so as to satisfy the relationship shown in 1).

Here, the difference between the vacuum level and the energy at the bottom of the conduction band (also referred to as electron affinity) is obtained by subtracting the energy gap from the difference between the vacuum level and the energy at the top of the valence band (also referred to as ionization potential). Value. The energy gap can be measured using a spectroscopic ellipsometer (HORIBA JOBIN YVON UT-300). The energy difference between the vacuum level and the upper end of the valence band can be measured using an ultraviolet photoelectron spectroscopy (UPS) apparatus (PHI VersaProbe).

FIG. 21A is a cross-sectional view of the oxide stack 404 having a band structure. FIG. 21B illustrates a band structure in the single-dot chain line G1-G2 of the oxide stack 404 illustrated in FIG. FIG. 21C illustrates a band structure along the dashed-dotted line G3-G4 in the oxide stack 404 illustrated in FIG. In FIGS. 21B and 21C, an insulating film with sufficiently large energy at the lower end of the conduction band is in contact with the first oxide layer 404a, the second oxide layer 404c, and the third oxide layer 404d. For example, a case where a silicon oxide film is provided will be described.

By selecting the first oxide layer 404a, the oxide semiconductor layer 404b, the second oxide layer 404c, and the third oxide layer 404d that satisfy the relationship of the mathematical formula (1), the oxide stack 404 can be made conductive. The first oxide layer 404a, the second oxide layer 404c, and the third oxide layer in which the energy at the lower end of the conduction band is higher than that of the oxide semiconductor layer 404b are used as the oxide semiconductor layer 404b having the lowest energy at the lower end of the band. A band structure surrounded by 404d is formed (see FIG. 21B).

Further, between the first oxide layer 404a and the oxide semiconductor layer 404b, between the oxide semiconductor layer 404b and the second oxide layer 404c, and between the oxide semiconductor layer 404b and the third oxide layer 404d. The energy at the lower end of the conduction band changes continuously. That is, there are almost no levels at these interfaces.

Therefore, in the oxide stack 404 having the band structure, electrons move mainly in the oxide semiconductor layer 404b. Therefore, even if a level exists at the interface with the insulating film, which is outside the oxide stack 404, the level hardly affects the movement of electrons. In addition, since there are almost no levels between the layers included in the oxide stack 404 or there is almost no level, movement of electrons in the region is not hindered. Therefore, the oxide semiconductor layer 404b in the oxide stack 404 has high electron mobility.

An oxide layer 404 in which one or more metal elements included in the oxide semiconductor layer 404 are stacked with the oxide semiconductor layer 404b interposed therebetween, that is, the oxide stack 404 stacked using a common main component, each layer is simply Rather than stacking, continuous junction as shown in FIGS. 21A and 21B (here, in particular, a U-shaped well structure in which the energy at the bottom of the conduction band changes continuously between the layers). Is formed so as to form. If the interface of each layer contains impurities such as trap centers and recombination centers for oxide semiconductors, or impurities that form barriers that hinder the flow of carriers, the continuity of the energy band is lost. This is because carriers disappear at the interface due to trapping or recombination.

In order to form a continuous bond, it is necessary to use a multi-chamber type film forming apparatus (sputtering apparatus) having a load lock chamber to successively laminate each layer without exposure to the atmosphere. Each chamber in the sputtering apparatus is evacuated (1 × 10 ⁻⁴ Pa to 5 ×) using an adsorption-type evacuation pump such as a cryopump so as to remove as much water as possible from the oxide semiconductor. It is preferable that it is about 10 ⁻⁷ Pa). Alternatively, it is preferable to combine a turbo molecular pump and a cold trap so that gas does not flow backward from the exhaust system into the chamber.

In order to obtain a high-purity intrinsic oxide semiconductor, it is necessary not only to evacuate the chamber to a high vacuum but also to increase the purity of the sputtering gas. Oxygen gas or argon gas used as a sputtering gas has a dew point of −40 ° C. or lower, preferably −80 ° C. or lower, more preferably −100 ° C. or lower. Ingestion can be prevented as much as possible.

Note that as shown in FIG. 22, trap levels caused by impurities and defects can be formed in the vicinity of the interface between the first oxide layer 404a and the second oxide layer 404c and the insulating film. The presence of the first oxide layer 404a and the second oxide layer 404c can keep the oxide semiconductor layer 404b away from the trap level. However, in the case where the energy difference between EcS1 or EcS3 and EcS2 is small, electrons in the oxide semiconductor layer 404b may reach the trap level beyond the first oxide layer 404a or the second oxide layer 404c. . By trapping electrons in the trap level, negative fixed charges are generated, and the threshold voltage of the transistor is shifted in the positive direction.

Similarly, although trap levels due to impurities and defects can be formed in the vicinity of the interface between the third oxide layer 404d and the insulating film, the oxide layer 404d can be formed by having the third oxide layer 404d. The semiconductor layer 404b and the trap level can be separated from each other. Note that in the case where the energy difference between EcS4 and EcS2 is small, electrons in the oxide semiconductor layer 404b may reach the trap level beyond the third oxide layer 404d. By trapping electrons in the trap level, negative fixed charges are generated, and the threshold voltage of the transistor is shifted in the positive direction.

Therefore, if the energy difference between EcS1, EcS3, EcS4, and EcS2 is 0.1 eV or more, preferably 0.15 eV or more, variation in the threshold voltage of the transistor is reduced and stable electrical characteristics are obtained. ,preferable.

<Deposition of oxide stack>
Each oxide layer included in the multilayer structure contains at least indium (In) and is formed using a sputtering target that can be formed by a sputtering method, preferably a DC sputtering method. Since the conductivity increases by including indium in the sputtering target, it is easy to form a film by a DC sputtering method.

The material forming the first oxide layer 404a and the second oxide layer 404c is an In-M-Zn oxide (such as Al, Ti, Ga, Ge, Y, Zr, Sn, La, Ce, or Hf). A material represented by (metal) is used. As M, Ga is preferably used. However, if the ratio of Ga to be included is large, specifically, it is a material that can be expressed as InGa _X Zn _Y O _Z and Y = 10, powder (particles) may be generated during film formation, and film formation is performed by sputtering. It is difficult to do so.

Note that the first oxide layer 404a and the second oxide layer 404c are each formed using a material with a lower atomic ratio of indium than the material used for the oxide semiconductor layer 404b. The contents of indium, gallium, and the like in the oxide layer can be compared by time-of-flight secondary ion mass spectrometry (TOF-SIMS) or X-ray electron spectroscopy (XPS).

The first oxide layer 404a may have an amorphous structure by containing a constituent element (eg, silicon) of the base insulating layer 402 as an impurity. Note that the oxide semiconductor layer 404b forming the channel preferably includes a crystal part. In the case where the oxide semiconductor layer 404b having a crystal part is stacked over the first oxide layer 404a having an amorphous structure, the oxide stack can be referred to as a heterostructure having a different crystal structure.

The second oxide layer 404c may have an amorphous structure or a crystal part. Note that when the second oxide layer 404c is formed over the oxide semiconductor layer 404b having a crystal part, the second oxide layer 404c is likely to be a film having a crystal structure. In some cases, it may be difficult to determine the boundary between the layer 404b and the second oxide layer 404c by TEM (Transmission Electron Microscope) observation. Note that since the crystallinity of the second oxide layer 404c is lower than that of the oxide semiconductor layer 404b, it can be said that the boundary can be determined based on the degree of crystallinity.

Note that in the oxide stack 404, at least the oxide semiconductor layer 404b is preferably a CAAC-OS (C Axis Crystalline Oxide Semiconductor) layer. In this specification and the like, a CAAC-OS layer refers to an oxide semiconductor film including a crystal part whose c-axis is substantially perpendicular to the surface of the oxide semiconductor layer.

The CAAC-OS layer is not completely single crystal nor completely amorphous. The CAAC-OS layer is an oxide semiconductor layer with a crystal-amorphous mixed phase structure where crystal parts and amorphous parts are included in an amorphous phase. Note that the crystal part is often large enough to fit in a cube whose one side is less than 100 nm. Further, in the observation image by the transmission electron microscope, the boundary between the amorphous part and the crystal part included in the CAAC-OS layer is not clear. Further, a grain boundary (also referred to as a grain boundary) cannot be confirmed in the CAAC-OS layer by TEM. Therefore, in the CAAC-OS layer, reduction in electron mobility due to grain boundaries is suppressed.

In the crystal part included in the CAAC-OS layer, the c-axis is aligned in a direction parallel to the normal vector of the formation surface of the CAAC-OS layer or the normal vector of the surface. Note that the directions of the a-axis and the b-axis may be different between different crystal parts. In this specification and the like, a simple term “vertical” includes a range from 85 ° to 95 °. In addition, a simple term “parallel” includes a range from −5 ° to 5 °.

Note that the distribution of crystal parts in the CAAC-OS layer is not necessarily uniform. For example, in the formation process of the CAAC-OS layer, in the case where crystal growth is performed from the surface side of the oxide semiconductor layer, the ratio of crystal parts in the vicinity of the surface is higher in the vicinity of the formation surface. In addition, when oxygen is added to the CAAC-OS layer, the crystal part in a region to which the element or oxygen is added becomes amorphous in some cases.

Since the c-axis of the crystal part included in the CAAC-OS layer is aligned in a direction parallel to the normal vector of the formation surface of the CAAC-OS layer or the normal vector of the surface, the shape of the CAAC-OS layer (formation surface) Depending on the cross-sectional shape of the surface or the cross-sectional shape of the surface). Note that the c-axis direction of the crystal part is parallel to the normal vector of the surface where the CAAC-OS layer is formed or the normal vector of the surface. The crystal part is formed by film formation or by performing crystallization treatment such as heat treatment after film formation.

In a transistor using a CAAC-OS film, change in electrical characteristics due to irradiation with visible light or ultraviolet light is small. Therefore, the transistor has high reliability.

Note that in the oxide stack 404, it is preferable that the first oxide layer 404a have an amorphous structure, and a CAAC-OS film be formed from the surface of the amorphous structure to form the oxide semiconductor layer 404b.

<Method for Forming CAAC-OS Film>
For example, the CAAC-OS film is formed by a sputtering method using a polycrystalline oxide semiconductor sputtering target. When ions collide with the sputtering target, the crystal region included in the sputtering target is cleaved from the ab plane, and may be separated as flat or pellet-like sputtering particles having a plane parallel to the ab plane. is there. In this case, the flat-plate-like sputtered particle reaches the substrate while maintaining a crystalline state, whereby a CAAC-OS film can be formed.

The flat sputtered particles have, for example, a circle-equivalent diameter of a plane parallel to the ab plane of 3 nm to 10 nm and a thickness (a length in a direction perpendicular to the ab plane) of 0.7 nm to less than 1 nm. is there. The flat sputtered particles may have a regular triangle or regular hexagonal plane parallel to the ab plane. Here, the equivalent-circle diameter of a surface means the diameter of a perfect circle that is equal to the area of the surface.

In order to form the CAAC-OS film, the following conditions are preferably applied.

By increasing the substrate temperature during film formation, migration of sputtered particles occurs after reaching the substrate. Specifically, the deposition is performed at a substrate temperature of 100 ° C. to 740 ° C., preferably 200 ° C. to 500 ° C. By increasing the substrate temperature during film formation, when the flat sputtered particles reach the substrate, migration occurs on the substrate, and the flat surface of the sputtered particles adheres to the substrate. At this time, since the sputtered particles are positively charged and the sputtered particles adhere to the substrate while being repelled, the sputtered particles are not biased and do not overlap unevenly, and a CAAC-OS film having a uniform thickness is formed. Can be membrane.

By reducing the mixing of impurities during film formation, the crystal state can be prevented from being broken by impurities. For example, the concentration of impurities (such as hydrogen, water, carbon dioxide, and nitrogen) existing in the deposition chamber may be reduced. Further, the impurity concentration in the deposition gas may be reduced. Specifically, a deposition gas having a dew point of −80 ° C. or lower, preferably −100 ° C. or lower is used.

In addition, it is preferable to reduce plasma damage during film formation by increasing the oxygen ratio in the film formation gas and optimizing electric power. The oxygen ratio in the deposition gas is 30% by volume or more, preferably 100% by volume.

Heat treatment may be performed after the CAAC-OS film is formed. The temperature of the heat treatment is 100 ° C. or higher and 740 ° C. or lower, preferably 200 ° C. or higher and 500 ° C. or lower. The heat treatment time is 1 minute to 24 hours, preferably 6 minutes to 4 hours. Further, the heat treatment may be performed in an inert atmosphere or an oxidizing atmosphere. Preferably, after heat treatment in an inert atmosphere, heat treatment is performed in an oxidizing atmosphere. By the heat treatment in an inert atmosphere, the impurity concentration of the CAAC-OS film can be reduced in a short time. On the other hand, oxygen vacancies may be generated in the CAAC-OS film by heat treatment in an inert atmosphere. In that case, the oxygen vacancies can be reduced by heat treatment in an oxidizing atmosphere. Further, by performing heat treatment, the crystallinity of the CAAC-OS film can be further increased. Note that the heat treatment may be performed under a reduced pressure of 1000 Pa or less, 100 Pa or less, 10 Pa or less, or 1 Pa or less. Under reduced pressure, the impurity concentration of the CAAC-OS film can be further reduced in a short time.

As an example of the sputtering target, an In—Ga—Zn—O compound target is described below.

In-Ga-Zn which is polycrystalline by mixing InO _X powder, GaO _Y powder and ZnO _Z powder at a predetermined molar ratio, and after heat treatment at a temperature of 1000 ° C to 1500 ° C. -O compound target. X, Y, and Z are arbitrary positive numbers. Here, the predetermined mole number ratio is, for example, 2: 2: 1, 8: 4: 3, 3: 1: 1, 1: 1: 1, 4 for InO _X powder, GaO _Y powder, and ZnO _Z powder. : 2: 3 or 3: 1: 2. In addition, what is necessary is just to change suitably the kind of powder, and the mol number ratio to mix with the sputtering target to produce.

Alternatively, the CAAC-OS film is formed by the following method.

First, the first oxide semiconductor film is formed with a thickness greater than or equal to 1 nm and less than 10 nm. The first oxide semiconductor film is formed by a sputtering method. Specifically, the film formation is performed at a substrate temperature of 100 ° C. or higher and 500 ° C. or lower, preferably 150 ° C. or higher and 450 ° C. or lower, and an oxygen ratio in the film forming gas is 30% by volume or higher, preferably 100% by volume.

Next, heat treatment is performed so that the first oxide semiconductor film becomes a first CAAC-OS film with high crystallinity. The temperature of the heat treatment is 350 ° C to 740 ° C, preferably 450 ° C to 650 ° C. The heat treatment time is 1 minute to 24 hours, preferably 6 minutes to 4 hours. Further, the heat treatment may be performed in an inert atmosphere or an oxidizing atmosphere. Preferably, after heat treatment in an inert atmosphere, heat treatment is performed in an oxidizing atmosphere. By the heat treatment in the inert atmosphere, the impurity concentration of the first oxide semiconductor film can be reduced in a short time. On the other hand, oxygen vacancies may be generated in the first oxide semiconductor film by heat treatment in an inert atmosphere. In that case, the oxygen vacancies can be reduced by heat treatment in an oxidizing atmosphere. Note that the heat treatment may be performed under a reduced pressure of 1000 Pa or less, 100 Pa or less, 10 Pa or less, or 1 Pa or less. Under reduced pressure, the impurity concentration of the first oxide semiconductor film can be further reduced in a short time.

When the thickness of the first oxide semiconductor film is greater than or equal to 1 nm and less than 10 nm, the first oxide semiconductor film can be easily crystallized by heat treatment as compared with the case where the thickness is greater than or equal to 10 nm.

Next, a second oxide semiconductor film having the same composition as the first oxide semiconductor film is formed to a thickness of greater than or equal to 10 nm and less than or equal to 50 nm. The second oxide semiconductor film is formed by a sputtering method. Specifically, the film formation is performed at a substrate temperature of 100 ° C. or higher and 500 ° C. or lower, preferably 150 ° C. or higher and 450 ° C. or lower, and an oxygen ratio in the film forming gas is 30% by volume or higher, preferably 100% by volume.

Next, heat treatment is performed, and the second oxide semiconductor film is solid-phase grown from the first CAAC-OS film, whereby the second CAAC-OS film with high crystallinity is obtained. The temperature of the heat treatment is 350 ° C to 740 ° C, preferably 450 ° C to 650 ° C. The heat treatment time is 1 minute to 24 hours, preferably 6 minutes to 4 hours. Further, the heat treatment may be performed in an inert atmosphere or an oxidizing atmosphere. Preferably, after heat treatment in an inert atmosphere, heat treatment is performed in an oxidizing atmosphere. By the heat treatment in the inert atmosphere, the impurity concentration of the second oxide semiconductor film can be reduced in a short time. On the other hand, oxygen vacancies may be generated in the second oxide semiconductor film by heat treatment in an inert atmosphere. In that case, the oxygen vacancies can be reduced by heat treatment in an oxidizing atmosphere. Note that the heat treatment may be performed under a reduced pressure of 1000 Pa or less, 100 Pa or less, 10 Pa or less, or 1 Pa or less. Under reduced pressure, the impurity concentration of the second oxide semiconductor film can be further reduced in a short time.

As described above, a CAAC-OS film with a total thickness of 10 nm or more can be formed. The CAAC-OS film can be favorably used as the oxide semiconductor layer in the oxide stack.

The structures, methods, and the like described in this embodiment can be combined as appropriate with any of the structures, methods, and the like described in the other embodiments.

(Embodiment 2)
In this embodiment, one embodiment of a semiconductor device including the stacked structure described in Embodiment 1 and a method for manufacturing the semiconductor device will be described with reference to FIGS. In this embodiment, a top-gate transistor including an oxide semiconductor layer is described as an example of a semiconductor device.

<Configuration Example 1 of Semiconductor Device>
FIG. 1 illustrates a configuration example of the transistor 310. 1A is a plan view of the transistor 310, FIG. 1B is a cross-sectional view taken along line X1-Y1 in FIG. 1A, and FIG. 1C is a cross-sectional view along V1- in FIG. It is sectional drawing in W1. FIG. 1D is a partially enlarged view of a region 200 in FIG. Note that in FIG. 1A, some components (eg, the protective insulating layer 414) are not illustrated in order to avoid complexity.

A transistor 310 illustrated in FIG. 1 includes a base insulating layer 402 formed over a substrate 400, an island-shaped oxide stack 404 formed over the base insulating layer, a part of the top surface of the island-shaped oxide stack, and channel formation. Provided on the first source electrode layer 406a and the first drain electrode layer 406b and the first source electrode layer 406a and the first drain electrode layer 406b, which are in contact with the side surfaces in the direction, respectively. The second source electrode layer 408a and the second drain electrode layer 408b made of a metal nitride film, which are in contact with a part thereof, are provided over the second source electrode layer 408a and the second drain electrode layer 408b. A gate insulating layer 410 in contact with the upper surface of the oxide stack 404 between the source electrode layer 408a and the second drain electrode layer 408b, and the oxide product through the gate insulating layer 410 And a gate electrode layer 412 which overlaps with the 404, the protective insulating layer 414 provided over and in contact with the gate insulating layer 410 and the gate electrode layer 412. Note that another insulating layer may be formed over the protective insulating layer 414.

<< Board >>
The substrate 400 is not limited to a simple support material, and may be a substrate on which other devices such as transistors are formed. In this case, at least one of the gate electrode layer 412, the first source electrode layer 406a, the first drain electrode layer 406b, the second source electrode layer 408a, and the second drain electrode layer 408b of the transistor 310 It may be electrically connected to other devices.

≪Base insulation layer≫
The base insulating layer 402 serves to prevent diffusion of impurities from the substrate 400 and also serves to supply oxygen to the oxide stack 404. Therefore, an insulating layer containing oxygen is used. In addition, when the substrate 400 is a substrate on which another device is formed as described above, the base insulating layer 402 also has a function as an interlayer insulating film. In that case, it is preferable to perform a planarization process by a CMP (Chemical Mechanical Polishing) method or the like so that the surface becomes flat.

In the transistor 310 of this embodiment, the base insulating layer 402 containing oxygen is provided below a stacked structure (oxide stack 404) including an oxide semiconductor layer. With such a structure, oxygen contained in the base insulating layer 402 can be supplied to the channel formation region. The base insulating layer 402 preferably includes a region containing oxygen in excess of the stoichiometric composition. When the base insulating layer 402 contains excessive oxygen, supply of oxygen to the channel formation region can be further promoted.

Note that in this specification and the like, excess oxygen means oxygen that can move in an oxide semiconductor layer, silicon oxide, or silicon oxynitride by heat treatment, or oxygen in an original stoichiometric ratio. Oxygen present in excess or oxygen having a function of filling or filling Vo (oxygen vacancy) due to lack of oxygen.

≪Gate insulation layer≫
Oxygen is supplied to the oxide stack 404 also from the gate insulating layer 410 provided in contact with the oxide stack 404 above. The gate insulating layer 410 has a region in contact with the base insulating layer 402 at the outer peripheral portion of the island-shaped oxide stack 404. Therefore, oxygen contained in the base insulating layer 402 can be supplied to the oxide stack 404 from the region where the gate insulating layer 410 and the base insulating layer 402 are in contact with each other using the gate insulating layer 410 as a path.

The gate insulating layer 410 is a layer serving as a path for supplying oxygen contained in the base insulating layer 402 to the oxide stack 404. The gate insulating layer 410 includes aluminum oxide, magnesium oxide, silicon oxide, silicon oxynitride, silicon nitride oxide, silicon nitride, gallium oxide, germanium oxide, yttrium oxide, zirconium oxide, lanthanum oxide, neodymium oxide, hafnium oxide, and tantalum oxide. Can be used. Further, the gate insulating layer 410 may be a stack of any of the above materials.

Since oxygen is supplied from the lower side and the upper side of the oxide stack 404, oxygen vacancies that can be included in the oxide stack 404 can be reduced.

≪Protective insulation layer≫
In the transistor 310, as the protective insulating layer 414 provided over the gate insulating layer 410 and the gate electrode layer 412, an insulating layer having a lower oxygen permeability (having a barrier property against oxygen) than the gate insulating layer 410 is provided. By providing the protective insulating layer 414 having a barrier property against oxygen over and in contact with the gate insulating layer 410 and the gate electrode layer 412, release of oxygen from the gate insulating layer 410 can be suppressed. Since the gate insulating layer 410 is an insulating layer serving as a path for supplying oxygen to the channel formation region, oxygen desorption from the gate insulating layer 410 is suppressed by suppressing release of oxygen from the gate insulating layer 410. In addition, extraction of oxygen from the oxide stack 404 can be suppressed, and as a result, oxygen vacancies in the channel formation region can be suppressed. As such a protective insulating layer, for example, a silicon nitride film or a silicon nitride oxide film can be provided.

In addition to oxygen vacancies, hydrogen serves as a carrier supply source in the oxide semiconductor. When hydrogen is contained in the oxide semiconductor, a donor is generated at a shallow level from the conduction band, and the resistance is reduced (n-type). Therefore, it is preferable to reduce the concentration of hydrogen contained in the protective insulating layer 414. Specifically, the hydrogen concentration contained in the protective insulating layer 414 is preferably less than 5 × 10 ¹⁹ cm ⁻³ and more preferably less than 5 × 10 ¹⁸ cm ⁻³ .

≪Oxide stack≫
The oxide stack 404 includes at least an oxide semiconductor layer 404b that forms a channel, a first oxide layer 404a provided between the oxide semiconductor layer 404b and the base insulating layer 402, and an oxide semiconductor layer 404b. A second oxide layer 404c provided between the gate insulating layer 410 and a first oxide layer 404a, an oxide semiconductor layer 404b, and a second oxide layer 404c provided in contact with the side surfaces of the second oxide layer 404c. 3 oxide layers 404d. The third oxide layer 404d has a curved surface.

Note that as described in Embodiment 1, the third oxide layer 404d is formed using a reaction product of an oxide film to be the first oxide layer 404a. Therefore, the boundary between the first oxide layer 404a and the third oxide layer 404d may not be clear. In addition, the second oxide layer 404c may be indistinguishable from the third oxide layer 404d in the case where the structure is similar to that of the first oxide layer 404a. In FIG. 1D, which is an enlarged view of the region 200, the boundary is illustrated by a dotted line for convenience.

Here, the cross section of the oxide stack 404 has a curved surface by including the third oxide layer 404d having a curved surface. Accordingly, coverage with a film formed over the oxide stack 404 (eg, a conductive film or a gate insulating film included in the source electrode layer and the drain electrode layer) can be improved. Accordingly, a film formed over the oxide stack 404 can be formed uniformly, and an impurity element enters the oxide stack 404 from a region with a low film density or a region where no film is formed, so that the semiconductor device Deterioration of characteristics can be suppressed.

Further, as illustrated in FIG. 1C, by including the third oxide layer 404d, the distance between the oxide semiconductor layer 404b forming the channel and the gate electrode layer 412 can be increased in the channel width direction. Can do. Thus, generation of parasitic channels that can occur in the channel width direction of the island-shaped oxide stack 404 can be suppressed.

The first oxide layer 404a and the second oxide layer 404c are oxide layers containing one or more metal elements included in the oxide semiconductor layer 404b. Embodiment 1 can be referred to for details of the oxide stack 404.

In the oxide stack 404, an oxide layer that has the same main component as the oxide semiconductor layer 404b and less oxygen vacancies than the oxide semiconductor layer 404b so as to surround the oxide semiconductor layer 404b that forms a channel. By providing (the first oxide layer 404a, the second oxide semiconductor layer 404b, and the third oxide layer 404d), formation of oxygen vacancies in the channel of the transistor can be suppressed.

Note that in order to make the oxide semiconductor layer intrinsic or substantially intrinsic, the silicon concentration is less than 1 × 10 ¹⁹ atoms / cm ³ , preferably less than 5 × 10 ¹⁸ atoms / cm ³ , preferably in SIMS analysis. Is less than 3 × 10 ¹⁸ atoms / cm ³ , preferably less than 1 × 10 ¹⁸ atoms / cm ³ . The hydrogen concentration is 2 × 10 ²⁰ atoms / cm ³ or less, preferably 5 × 10 ¹⁹ atoms / cm ³ or less, more preferably 1 × 10 ¹⁹ atoms / cm ³ or less, and even more preferably 5 × 10 ¹⁸ atoms / cm ^3. cm ³ or less. The nitrogen concentration is less than 5 × 10 ¹⁹ atoms / cm ³ , preferably 5 × 10 ¹⁸ atoms / cm ³ or less, more preferably 1 × 10 ¹⁸ atoms / cm ³ or less, and further preferably 5 × 10 ¹⁷ atoms / cm ^3. cm ³ or less.

In addition, in the case where the oxide semiconductor layer includes a crystal, the crystallinity of the oxide semiconductor layer may be reduced if silicon or carbon is included at a high concentration. In order not to reduce the crystallinity of the oxide semiconductor layer, the silicon concentration is less than 1 × 10 ¹⁹ atoms / cm ³ , preferably less than 5 × 10 ¹⁸ atoms / cm ³ , more preferably 1 × 10 ¹⁸ atoms / cm ^3. It may be less than. The carbon concentration may be less than 1 × 10 ¹⁹ atoms / cm ³ , preferably less than 5 × 10 ¹⁸ atoms / cm ³ , and more preferably less than 1 × 10 ¹⁸ atoms / cm ³ .

In addition, the off-state current of the transistor in which the oxide semiconductor film purified as described above is used for a channel formation region is extremely small. For example, the drain current when the transistor including the highly purified oxide semiconductor film is off is 1 × 10 ⁻¹⁸ A or less, preferably 1 × 10 ⁻²¹ A or less at room temperature (about 25 ° C.). More preferably, it is 1 × 10 ⁻²⁴ A or less, or 1 × 10 ⁻¹⁵ A or less, preferably 1 × 10 ⁻¹⁸ A or less, more preferably 1 × 10 ⁻²¹ A or less at 85 ° C. . Note that an off state of a transistor means a state where a gate voltage is sufficiently lower than a threshold voltage in the case of an n-channel transistor. Specifically, when the gate voltage is 1 V or higher, 2 V or higher, or 3 V or lower than the threshold voltage, the transistor is turned off.

≪Source electrode layer and drain electrode layer≫
In the transistor 310, the source electrode layer and the drain electrode layer are in contact with the side surface in the channel length direction of the oxide stack 404, the first source electrode layer 406a and the first drain electrode layer 406b, the first source electrode layer 406a, and the first source electrode layer 406a. The second source electrode layer 408a and the second drain provided on the first drain electrode layer 406b and having a region extending in the channel length direction from the first source electrode layer 406a and the first drain electrode layer 406b. An electrode layer 408b is provided.

The first source electrode layer 406a and the first drain electrode layer 406b can be formed using a conductive material that is easily bonded to oxygen. For example, Al, Cr, Cu, Ta, Ti, Mo, W, etc. can be used. It is particularly preferable to use W having a high melting point because the subsequent process temperature can be made relatively high. Note that the conductive material that easily binds to oxygen includes a material that easily diffuses oxygen.

When such a conductive material is brought into contact with the oxide stack 404, oxygen in the oxide stack 404 is taken into the conductive material side that is easily bonded to oxygen. Since there are several heating steps in the manufacturing process of the transistor, a region in the vicinity of the interface in contact with the first source electrode layer 406a and the first drain electrode layer 406b in the oxide stack 404 due to the movement of oxygen. Oxygen vacancies are generated and an n-type region 405 is formed (see FIG. 1D). Accordingly, the n-type region 405 can serve as a source or a drain of the transistor 310.

Note that the constituent elements of the first source electrode layer 406a and the first drain electrode layer 406b may be mixed in the region 405. In the first source electrode layer 406a and the first drain electrode layer 406b in contact with the region 405, a region where oxygen concentration is partly high can be formed. Further, in the first source electrode layer 406a and the first drain electrode layer 406b in contact with the region 405, a constituent element of the oxide stack 404 may be mixed. In other words, in the vicinity of the interface in contact with the first source electrode layer 406a and the first drain electrode layer 406b in the oxide stack 404, a portion that can be referred to as a mixed region or a mixed layer of the two contacted layers is formed. Sometimes. In FIG. 1, the interface of the n-type region 405 is schematically shown by a dotted line. The same applies to the subsequent drawings.

In the case of forming a transistor with an extremely short channel length, the n-type region due to the generation of oxygen vacancies may extend in the channel length direction of the transistor. In this case, a state (conduction state) in which on / off cannot be controlled by a threshold voltage shift or a gate voltage appears in the electrical characteristics of the transistor. Therefore, in the case of forming a transistor with an extremely short channel length, it is not preferable to use a conductive material that easily binds to oxygen for the source electrode and the drain electrode.

Therefore, the second source electrode layer 408a and the second drain electrode layer 408b having a region extending in the channel length direction from the first source electrode layer 406a and the first drain electrode layer 406b are used as the first source electrode. The second source electrode layer 408a and the second drain electrode layer 408b which are stacked over the layer 406a and the second drain electrode 406b and define a channel length are formed using a conductive material which is not easily bonded to oxygen. For example, tantalum nitride, titanium nitride, or the like is preferably used as the conductive material. Note that the conductive material which is difficult to bond with oxygen includes a material which hardly diffuses oxygen.

Note that in the transistor having the structure in FIG. 1, the channel length refers to the distance L2 between the second source electrode layer 408a and the second drain electrode layer 408b.

In the transistor having the structure in FIG. 1, a channel refers to the oxide semiconductor layer 404b between the second source electrode layer 408a and the second drain electrode layer 408b.

In the transistor having the structure in FIG. 1, the channel formation region refers to the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer between the second source electrode layer 408a and the second drain electrode layer 408b. This refers to the oxide layer 404c.

By using the conductive material which is not easily bonded to oxygen for the second source electrode layer 408a and the second drain electrode layer 408b, formation of oxygen vacancies in the channel formation region formed in the oxide stack 404 is suppressed. Thus, the n-type channel can be suppressed. Therefore, good electrical characteristics can be obtained even with a transistor having a very short channel length.

Note that, when the source electrode layer and the drain electrode layer are formed using only a conductive material that does not easily bond to oxygen, the contact resistance with the oxide stack 404 becomes too high, so that the first source electrode is formed as shown in FIG. The layer 406a and the first drain electrode layer 406b are formed over the oxide stack 404, and the second source electrode layer 408a and the second source electrode layer 408a are in contact with the first source electrode layer 406a and the first drain electrode layer 406b. The drain electrode layer 408b is preferably formed.

At this time, the contact area between the first source electrode layer 406a and the first drain electrode layer 406b and the oxide stack 404 is increased, and the contact resistance is reduced by the region 405 that is n-typed by generation of oxygen vacancies. The contact area between the electrode layer 408a and the second drain electrode layer 408b and the oxide stack 404 is preferably small. This is because if the contact area between the second source electrode layer 408a and the second drain electrode layer 408b and the oxide stack 404 is large, the electrical characteristics of the transistor may be deteriorated.

The distance L2 between the second source electrode layer 408a and the second drain electrode layer 408b can be set to a value smaller than L1, and for example, favorable transistor electrical characteristics can be obtained even when the thickness is 30 nm or less.

≪Gate electrode layer≫
The gate electrode layer 412 can be formed using a conductive film such as Al, Ti, Cr, Co, Ni, Cu, Y, Zr, Mo, Ru, Ag, Ta, and W.

The structure of the transistor of one embodiment of the present invention can suppress an increase in oxygen vacancies in the oxide semiconductor layer. Therefore, a semiconductor device that exhibits favorable electrical characteristics and high long-term reliability can be provided.

<Configuration Example 2 of Semiconductor Device>
FIG. 2 illustrates a configuration example of the transistor 320. 2A is a plan view of the transistor 320, FIG. 2B is a cross-sectional view taken along line X2-Y2 in FIG. 2A, and FIG. 2C is a cross-sectional view along V2- in FIG. It is sectional drawing in W2. Note that in FIG. 2A, some components (eg, the protective insulating layer 414) are not illustrated in order to avoid complexity.

2 includes a base insulating layer 402 formed over a substrate 400, an island-shaped oxide stack 404 formed over the base insulating layer, a part of the top surface of the island-shaped oxide stack, and channel formation. The upper surface of the oxide stack 404 is provided to cover the first source electrode layer 406a and the first drain electrode layer 406b in contact with the side surface in the direction, and the first source electrode layer 406a and the first drain electrode layer 406b. Are provided on the second source electrode layer 408a and the second drain electrode layer 408b made of a metal nitride film, and on the second source electrode layer 408a and the second drain electrode layer 408b. A gate insulating layer 410 in contact with the upper surface of the oxide stack 404 between the source electrode layer 408a and the second drain electrode layer 408b, and the oxide product through the gate insulating layer 410. And a gate electrode layer 412 which overlaps with the 404, the protective insulating layer 414 provided over and in contact with the gate insulating layer 410 and the gate electrode layer 412. Note that another insulating layer may be formed over the protective insulating layer 414.

One of the differences between the transistor 310 and the transistor 320 is the length L0 of the gate electrode layer 412 in the channel length direction. The transistor 320 has a structure in which the gate electrode layer 412 does not overlap with the first source electrode layer 406a and the first drain electrode layer 406b in order to reduce gate-drain and gate-source parasitic capacitance. When the length of the gate electrode layer 412 is L0, the parasitic capacitance between the gate and the drain and between the gate and the source can be reduced as much as possible by setting L1 ≧ L0 ≧ L2 (L0 is L2 or more and L1 or less). The frequency characteristics of the semiconductor device can be improved. Note that it is preferable that (L1−L2) / 2 be less than 20% or less of L2 in order to obtain favorable transistor electrical characteristics.

The upper area of the transistor 320 is preferably 1 to 25 μm ² or less. For example, in the transistor 320 in FIG. 2, when the length of the oxide stack 404 in the channel length direction is 300 nm, L0 is preferably 40 nm and L2 is preferably 30 nm. At this time, the upper area of the island-shaped oxide stack 404 can be 1 μm ² or less.

However, in a semiconductor device that does not require high frequency characteristics, L0 ≧ L1 ≧ L2 (L1 is not less than L2 and not more than L0) as shown in FIG. In such a structure, the difficulty of the process at the time of forming the gate electrode can be reduced.

Another difference between the transistor 310 and the transistor 320 is the structure of the gate electrode layer 412. In the transistor 320, the gate electrode layer 412 has a stacked structure of a first gate electrode layer 412a and a second gate electrode layer 412b in contact with the gate insulating layer 410. Here, as the first gate electrode layer 412a, a material similar to that of the second source electrode layer 408a and the second drain electrode layer 408b is used, whereby oxygen is extracted from the gate insulating layer 410 by the gate electrode layer 412. Can be prevented.

Another difference between the transistor 310 and the transistor 320 is the structure of the second source electrode layer 408a and the second drain electrode layer 408b. In the transistor 320, the width of the first source electrode layer 406a (or the first drain electrode layer 406b) in the channel width direction is W1, and the channel of the second source electrode layer 408a (or the second drain electrode layer 408b) is used. When the width in the width direction is W2, by setting W1> W2, the second source electrode layer 408a (or the second drain electrode layer 408b) becomes the first source electrode layer 406a (or the first drain). The electrode layer 406b) is covered.

With such a structure, the gate insulating layer 410 can be configured not to be in contact with the first source electrode layer 406a and the first drain electrode layer 406b which are formed using a conductive material which is easily bonded to oxygen. . Accordingly, extraction of oxygen from the gate insulating layer 410 can be prevented.

Since the gate insulating layer 410 is a layer serving as a supply path of oxygen to the oxide stack 404, oxygen extraction from the gate insulating layer 410 by the first source electrode layer 406a and the first drain electrode layer 406b is prevented. Thus, oxygen can be effectively supplied to the oxide stack 404. Accordingly, generation of oxygen vacancies in the oxide stack 404 can be suppressed, so that the reliability of the transistor 320 can be improved.

Note that the other structures of the transistor 320 are similar to those of the above-described transistor 310, and the description of the transistor 310 can be referred to.

<Configuration Example 3 of Semiconductor Device>
3A and 3B illustrate a modification example of the transistor 310 or the transistor 320 described above. A transistor 330 illustrated in FIG. 3A has a structure in which end portions of the first source electrode layer 406a and the first drain electrode layer 406b of the transistor 310 are different, and the transistor 340 illustrated in FIG. 320, the first source electrode layer 406a and the first drain electrode layer 406b have different end portions. Note that the structures other than the end portions of the first source electrode layer 406a and the first drain electrode layer 406b are similar to those of the above transistor, and the above description can be referred to.

3C and 3D are enlarged views of the first source electrode layer 406a in the transistor 330 and the transistor 340. FIG. In the transistor 330 and the transistor 340, end portions of the first source electrode layer 406 a and the first drain electrode layer 406 b have a plurality of steps in a step shape, and a lower end surface of the oxide stack 404 The upper end surface is preferably formed at an angle of θ1 with respect to the upper surface, and the upper end face is preferably formed at an angle of θ2 with respect to the lower upper surface. Further, a curved surface having a radius of curvature R1 is provided between the lower upper surface and the lower end surface, and a curved surface having a radius of curvature R3 is provided between the upper upper surface and the upper end surface, and between the lower upper surface and the upper end surface. Is preferably formed of a curved surface having a radius of curvature of R2.

Note that although FIG. 3 illustrates a shape in which two steps are provided at the ends of the first source electrode layer 406a and the first drain electrode layer 406b, the number of steps may be three or more. . It is preferable to increase the number of steps as the thickness of the first source electrode layer 406a and the first drain electrode layer 406b increases. Note that the end portions of the first source electrode layer 406a and the first drain electrode layer 406b are not necessarily targeted. In the case where the thickness of the first source electrode layer 406a and the first drain electrode layer 406b can be reduced, the number of steps may be one, that is, a shape having only θ1 and R1.

Here, θ1 and θ2 are 20 ° to 80 °, preferably 25 ° to 70 °, more preferably 30 ° to 60 °. R1, R2, and R3 are 10% or more and 100% or less of the film thickness based on the film thickness of the region where the first source electrode layer 406a or the first drain electrode layer 406b is not formed, Preferably they are 20% or more and 75% or less, More preferably, they are 30% or more and 60% or less. R2 is preferably larger than either R1 or R3, or both.

By forming the first source electrode layer 406a and the first drain electrode layer 406b in a shape having a plurality of steps as described above, a film formed above them, specifically, a second source The coverage of the electrode layer 408a, the second drain electrode layer 408b, the gate insulating layer 410, and the like can be improved, so that the electrical characteristics and long-term reliability of the transistor can be improved.

In addition, the end surface of the second source electrode layer 408 a or the second drain electrode layer 408 b is formed at an angle of θ3 with respect to the upper surface of the oxide stack 404. θ3 is 30 ° to 80 °, preferably 35 ° to 75 °. With such an angle, the coverage of the gate insulating layer 410 and the like can be improved, and the electrical characteristics and long-term reliability of the transistor can be improved.

The above is the transistor of one embodiment of the present invention. In the transistor described in this embodiment, oxygen contained in the base insulating layer is supplied to the channel formation region from the base insulating layer or through the gate insulating layer that is in contact with the base insulating layer at the outer periphery of the island-shaped oxide stack. It is possible to continue to supply. With the structure of the transistor, an increase in oxygen vacancies in the oxide semiconductor layer can be suppressed, the impurity concentration can be reduced, and the oxide semiconductor layer can be highly purified and intrinsic.

Accordingly, fluctuations in electrical characteristics such as threshold voltage fluctuations that can be caused by oxygen vacancies in the channel formation region can be suppressed. Specifically, for example, normally-off of the threshold voltage can be stably achieved. Therefore, a semiconductor device that exhibits favorable electrical characteristics and high long-term reliability can be provided.

Note that the structures of the transistors described in this embodiment can be freely combined with each other. For example, the gate electrode layer of the transistor 310 may have a stacked structure represented by the transistor 320.

Note that this embodiment can be combined with any of the other embodiments or examples in this specification as appropriate.

(Embodiment 3)
In this embodiment, an example of a method for manufacturing the transistor described in Embodiment 1 will be described. In this embodiment, the case where the transistor 340 illustrated in FIG. 3B is manufactured is described as an example with reference to FIGS.

First, the base insulating layer 402 is formed over the substrate 400 having an insulating surface (see FIG. 4A).

The base insulating layer 402 has a thickness of 1 nm to 100 nm, and is formed by plasma CVD, sputtering, or the like using aluminum oxide, magnesium oxide, silicon oxide, silicon oxynitride, gallium oxide, germanium oxide, yttrium oxide, zirconium oxide, or oxide. An insulating film containing oxygen such as lanthanum, neodymium oxide, hafnium oxide, or tantalum oxide can be used.

The base insulating layer 402 is formed of a material containing oxygen which can serve as a supply source of oxygen to the oxide stack 404 at least in an upper layer in contact with the oxide stack 404 and the gate insulating layer 410 to be formed later. In addition, it is preferable to form a film containing excessive oxygen.

In order to make the base insulating layer 402 contain excessive oxygen, for example, the base insulating layer 402 may be formed in an oxygen atmosphere. Alternatively, oxygen may be introduced excessively into the base insulating layer 402 after film formation to contain oxygen in excess, or both means may be combined.

In this embodiment, oxygen 302 (including at least one of oxygen radicals, oxygen atoms, and oxygen ions) is introduced into the base insulating layer 402 after deposition, so that a region containing excess oxygen is formed. As a method for introducing oxygen, an ion implantation method, an ion doping method, a plasma immersion ion implantation method, a plasma treatment, or the like can be used.

A gas containing oxygen can be used for the oxygen introduction treatment. As the gas containing oxygen, oxygen, dinitrogen monoxide, nitrogen dioxide, carbon dioxide, carbon monoxide, or the like can be used. Further, in the oxygen introduction treatment, a gas containing oxygen may contain a rare gas.

Next, the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c are formed over the base insulating layer 402 by a sputtering method, a CVD method, an MBE method, an ALD method, or a PLD method. The oxide stack 404 is formed by selective etching (see FIG. 4B). Note that a heating step may be performed before etching. Embodiment 1 can be referred to for details of the formation method of the oxide stack 404.

The materials described in Embodiment 1 can be used for the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c.

For example, the first oxide layer 404a includes an In: Ga: Zn = 1: 3: 2 [atomic ratio] In—Ga—Zn oxide, In: Ga: Zn = 1: 6: 4 [atom]. Number ratio] In—Ga—Zn oxide, In: Ga: Zn = 1: 9: 6 [atomic number ratio] In—Ga—Zn oxide, or an oxide having a composition in the vicinity thereof is used. preferable.

For example, the oxide semiconductor layer 404b includes an In: Ga: Zn = 1: 1: 1 [atomic ratio] In—Ga—Zn oxide, In: Ga: Zn = 3: 1: 2 [atoms]. The number ratio] In—Ga—Zn oxide or an oxide having a composition in the vicinity thereof is preferably used.

For example, for the second oxide layer 404c, an In—Ga—Zn oxide with In: Ga: Zn = 1: 3: 2 [atomic ratio] or an oxide having a composition in the vicinity thereof is used. It is preferable.

Note that for example, the composition of an oxide in which the atomic ratio of In, Ga, and Zn is In: Ga: Zn = a: b: c (a + b + c = 1) has an atomic ratio of In: Ga: Zn = A: B: C (A + B + C = 1) is in the vicinity of the oxide composition, a, b, c are (a−A) ² + (b−B) ² + (c−C) ² ≦ r ² Satisfying. For example, r may be 0.05.

Note that the composition of each oxide layer is not limited to the above-described atomic ratio. Note that the oxide semiconductor layer 404b preferably contains more indium than the first oxide layer 404a and the second oxide layer 404c. In oxide semiconductors, s orbitals of heavy metals mainly contribute to carrier conduction, and by increasing the In content, more s orbitals overlap, so an oxide having a composition with more In than Ga is indium. Is higher in mobility than an oxide having a composition equal to or less than that of gallium. In addition, gallium has a larger energy generation density of oxygen vacancies than indium, and thus oxygen vacancies are less likely to occur, so that an oxide having a high gallium content has stable characteristics.

Therefore, a transistor with high mobility can be realized by using an oxide containing a large amount of indium for the oxide semiconductor layer 404b. In addition, when an oxide containing a large amount of gallium is used on the interface side with the insulating layer, the reliability of the transistor can be improved.

An oxide semiconductor that can be used as the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c preferably contains at least indium (In) or zinc (Zn). Or it is preferable that both In and Zn are included. In particular, when the oxide semiconductor layer 404b contains indium, carrier mobility of the transistor can be increased, and when zinc is contained, a CAAC-OS film is easily formed, which is preferable. In addition, in order to reduce variation in electrical characteristics of a transistor including an oxide semiconductor layer, a stabilizer is preferably included in addition to indium and zinc.

Examples of the stabilizer include gallium (Ga), tin (Sn), hafnium (Hf), aluminum (Al), and zirconium (Zr). Other stabilizers include lanthanoids such as lanthanum (La), cerium (Ce), praseodymium (Pr), neodymium (Nd), samarium (Sm), europium (Eu), gadolinium (Gd), terbium (Tb). ), Dysprosium (Dy), holmium (Ho), erbium (Er), thulium (Tm), ytterbium (Yb), lutetium (Lu), and the like.

For example, as an oxide semiconductor, indium oxide, tin oxide, zinc oxide, In—Zn oxide, Sn—Zn oxide, Al—Zn oxide, Zn—Mg oxide, Sn—Mg oxide, In—Mg oxide In-Ga oxide, In-Ga-Zn oxide, In-Al-Zn oxide, In-Sn-Zn oxide, Sn-Ga-Zn oxide, Al-Ga-Zn oxide, Sn- Al—Zn oxide, In—Hf—Zn oxide, In—La—Zn oxide, In—Ce—Zn oxide, In—Pr—Zn oxide, In—Nd—Zn oxide, In—Sm— Zn oxide, In-Eu-Zn oxide, In-Gd-Zn oxide, In-Tb-Zn oxide, In-Dy-Zn oxide, In-Ho-Zn oxide, In-Er-Zn oxide , In-Tm-Zn oxide, In-Yb-Zn oxidation In-Lu-Zn oxide, In-Sn-Ga-Zn oxide, In-Hf-Ga-Zn oxide, In-Al-Ga-Zn oxide, In-Sn-Al-Zn oxide, In -Sn-Hf-Zn oxide and In-Hf-Al-Zn oxide can be used.

Note that as described in detail in Embodiment 1, the materials of the first oxide layer 404a and the second oxide layer 404c are selected so that the electron affinity is higher than that of the oxide semiconductor layer 404b.

Note that a sputtering method is preferably used for forming the oxide stack. As the sputtering method, an RF sputtering method, a DC sputtering method, an AC sputtering method, or the like can be used. In particular, it is preferable to use a DC sputtering method because dust generated during film formation can be reduced and the film thickness distribution is uniform.

Note that oxygen may be introduced into the first oxide layer 404a after the first oxide layer 404a is formed and before the oxide semiconductor layer 404b is formed. By the oxygen introduction treatment, the first oxide layer 404a contains excess oxygen, and the excess oxygen can be supplied to the oxide semiconductor layer 404b by heat treatment in a subsequent film formation step.

Therefore, oxygen vacancies in the oxide semiconductor layer 404b can be further suppressed by the oxygen introduction treatment into the first oxide layer 404a.

Note that the first oxide layer 404a may be amorphized by oxygen introduction treatment. In the oxide stack 404, at least the oxide semiconductor layer 404b is preferably a CAAC-OS film. Therefore, the oxygen introduction treatment is preferably performed after the formation of the first oxide layer 404a and before the formation of the oxide semiconductor layer 404b.

It is preferable to perform the first heat treatment after the oxide stack 404 is formed. The first heat treatment may be performed at a temperature of 250 ° C. to 650 ° C., preferably 300 ° C. to 500 ° C., in an inert gas atmosphere, an atmosphere containing an oxidizing gas of 10 ppm or more, or a reduced pressure state. The atmosphere for the heat treatment may be an atmosphere containing 10 ppm or more of an oxidizing gas in order to supplement the desorbed oxygen after the heat treatment in an inert gas atmosphere. By the first heat treatment, the crystallinity of the oxide semiconductor layer 404b is increased, and further, from at least one of the base insulating layer 402, the first oxide layer 404a, the oxide semiconductor layer 404b, and the second oxide layer 404c. Impurities such as hydrogen and water can be removed. Note that a heating step may be performed before the etching for forming the oxide stack 404.

After that, a first conductive film to be the first source electrode layer 406a and the first drain electrode layer 406b is formed over the oxide stack 404. As the first conductive film, Al, Cr, Cu, Ta, Ti, Mo, W, or an alloy material containing these as a main component can be used. For example, a 100 nm tungsten film is formed by sputtering or the like.

Next, the first conductive film is etched so as to be divided over the oxide stack 404, so that the first source electrode layer 406a and the first drain electrode layer 406b are formed (see FIG. 4C). At this time, the end portions of the first source electrode layer 406a and the first drain electrode layer 406b are preferably formed in a step shape as illustrated. The edge portion can be formed by alternately performing a step of retracting the resist mask by ashing and an etching step a plurality of times.

Note that although not illustrated, a part of the oxide stack 404 (an exposed region) may be etched by overetching the first conductive film.

Next, a second conductive film to be the second source electrode layer 408a and the second drain electrode layer 408b is formed over the oxide stack 404, the first source electrode layer 406a, and the first drain electrode layer 406b. To do. As the second conductive film, a metal nitride film such as tantalum nitride or titanium nitride or an alloy material containing these as a main component can be used. For example, a 20 nm tantalum nitride film is formed by sputtering or the like.

Next, the second conductive film is etched so as to be divided over the oxide stack 404, so that the second source electrode layer 408a and the second drain electrode layer 408b are formed (see FIG. 5A). At this time, a part of the oxide stack 404 (more specifically, part of the second oxide layer 404c) may be etched. Note that although not illustrated, the second source electrode layer 408a and the second drain electrode layer 408b are formed in the base insulating layer 402 by an etching process for forming the second source electrode layer 408a and the second drain electrode layer 408b. In some cases, the exposed area is etched and the film thickness of the area is reduced.

Note that in the case where a transistor with a very short channel length (between the second source electrode layer 408a and the second drain electrode layer 408b) is formed, at least a region where the second conductive film is divided is exposed to an electron beam or the like. The resist mask is processed using a method suitable for thin line processing, and the region is etched by an etching process. Note that when a positive resist is used as the resist mask, an exposure region can be minimized and throughput can be improved. By using such a method, a transistor with a channel length of 30 nm or less can be formed.

Next, the gate insulating layer 410 is formed over the oxide stack 404, the second source electrode layer 408a, and the second drain electrode layer 408b (see FIG. 5B). The gate insulating layer 410 includes aluminum oxide, magnesium oxide, silicon oxide, silicon oxynitride, silicon nitride oxide, silicon nitride, gallium oxide, germanium oxide, yttrium oxide, zirconium oxide, lanthanum oxide, neodymium oxide, hafnium oxide, and tantalum oxide. Etc. can be used. Note that the gate insulating layer 410 may be a stack of any of the above materials. The gate insulating layer 410 can be formed by a sputtering method, a CVD method, an MBE method, an ALD method, a PLD method, or the like. In particular, the gate insulating layer 410 is preferably formed by a CVD method, preferably a plasma CVD method, because the coverage is good.

After the gate insulating layer 410 is formed, second heat treatment is preferably performed. By the second heat treatment, impurities such as water and hydrogen contained in the gate insulating layer 410 can be desorbed (dehydrated or dehydrogenated). The temperature of the second heat treatment is preferably 300 ° C to 400 ° C. The second heat treatment is preferably performed in an atmosphere containing an oxidizing gas of 10 ppm or more in order to supplement the desorbed oxygen after the heat treatment in an inert gas atmosphere. By the second heat treatment, impurities such as hydrogen and water can be removed from the gate insulating layer 410. Further, impurities such as hydrogen and water may be further removed from the oxide stack 404. In addition, oxygen can be supplied to the gate insulating layer 410 by performing heat treatment in an atmosphere containing an oxidizing gas.

Note that the second heat treatment is preferably performed continuously in the deposition chamber after the gate insulating layer 410 is formed. Alternatively, the heat treatment at the time of forming the gate insulating layer 410 can also serve as the second heat treatment.

The second heat treatment is performed in a state where the first source electrode layer 406a and the first drain electrode layer 406b are in contact with the oxide stack 404, so that the oxide stack 404 can be easily bonded to oxygen. Oxygen is taken into the one source electrode layer 406a and the first drain electrode layer 406b. Therefore, oxygen vacancies are generated in a region in the vicinity of the interface in contact with the first source electrode layer 406a and the first drain electrode layer 406b in the oxide stack 404, so that an n-type region 405 is formed. However, n-type conversion may not occur depending on the temperature of the second heat treatment.

Next, a conductive film to be the gate electrode layer 412 is formed over the gate insulating layer 410. The conductive film can be formed by a sputtering method or the like. Then, the conductive film is etched so as to overlap with the channel formation region, so that the gate electrode layer 412 is formed (see FIG. 5C).

In this embodiment, the stacked structure of the first gate electrode layer 412a using the same material as the second conductive film and the second gate electrode layer 412b using the same material as the first conductive film is used. A gate electrode layer 412 is formed.

Next, a protective insulating layer 414 is formed over the gate insulating layer 410 and the gate electrode layer 412 (see FIG. 5D). As the protective insulating layer 414, an insulating layer having a lower oxygen permeability (having a barrier property against oxygen) than the gate insulating layer 410 is provided. For example, a silicon nitride film or a silicon nitride oxide film can be provided as the protective insulating layer 414.

Note that the protective insulating layer 414 is preferably formed by a sputtering method because the concentration of contained hydrogen is preferably reduced. The concentration of hydrogen contained in the protective insulating layer 414 is preferably less than 5 × 10 ¹⁹ cm ⁻³ and more preferably less than 5 × 10 ¹⁸ cm ⁻³ .

After the protective insulating layer 414 is formed, third heat treatment is preferably performed. The temperature of the third heat treatment is preferably 350 ° C. or higher and 450 ° C. or lower. By the third heat treatment, oxygen is easily released from the base insulating layer 402 and the gate insulating layer 410, so that oxygen vacancies in the oxide stack 404 can be reduced.

In addition, by the third heat treatment, oxygen can move from the oxide stack 404 to the first source electrode layer 406a and the first drain electrode layer 406b, which are easily bonded to oxygen. Therefore, the n-type region 405 may be more n-type. Alternatively, even when the region near the interface is not n-typed by the second heat treatment, the region can be n-typed by the third heat treatment.

Through the above, the transistor 340 of this embodiment can be manufactured.

Note that this embodiment can be combined with any of the other embodiments or examples in this specification as appropriate.

(Embodiment 4)
As an example of the semiconductor device according to one embodiment of the present invention, FIG. 23A illustrates an example of a circuit diagram of a NOR circuit that is a logic circuit. FIG. 23B is a circuit diagram of a NAND circuit.

In the NOR circuit shown in FIG. 23A, the transistors 801 and 802 which are p-channel transistors are transistors using a semiconductor material (eg, silicon) other than an oxide semiconductor in a channel formation region, and are n-channel transistors. Transistors 803 and 804 which are transistors each include an oxide stack including an oxide semiconductor and have a structure similar to that of the transistor described in Embodiment 2.

A transistor using a semiconductor material such as silicon can easily operate at high speed. On the other hand, a transistor including an oxide semiconductor can hold charge for a long time due to its characteristics.

In order to reduce the size of the logic circuit, the transistors 803 and 804 which are n-channel transistors are preferably stacked over the transistors 801 and 802 which are p-channel transistors. For example, the transistors 801 and 802 can be formed using a single crystal silicon substrate, and the transistors 803 and 804 can be formed over the transistors 801 and 802 with an insulating layer interposed therebetween.

Note that in the NOR circuit illustrated in FIG. 23A, the transistors 803 and 804 have the same structure as the transistor 520, and the potential of the second gate electrode is controlled to be, for example, GND so that the transistor 803 , 804 can be made more positive, and a normally-off transistor can be obtained.

In the NAND circuit illustrated in FIG. 23B, the transistors 811 and 814 which are p-channel transistors are transistors each using a semiconductor material other than an oxide semiconductor (eg, silicon) in a channel formation region. Transistors 812 and 813 which are channel transistors each include an oxide stack including an oxide semiconductor and have a structure similar to that of the transistor described in Embodiment 2.

Note that in the NAND circuit illustrated in FIG. 23B, the transistors 812 and 813 have the same structure as the transistor 520, and the potential of the second gate electrode is controlled to be, for example, GND so that the transistor 812 , 813 can be made more positive, and a normally-off transistor can be obtained.

Similarly to the NOR circuit illustrated in FIG. 23A, in order to reduce the size of the logic circuit, the transistors 812 and 813 which are n-channel transistors are provided over the transistors 811 and 812 which are p-channel transistors. It is preferable to be laminated.

In the semiconductor device described in this embodiment, power consumption can be sufficiently reduced by using a transistor with an extremely small off-state current that uses an oxide semiconductor for a channel formation region.

In addition, a semiconductor device in which miniaturization and high integration are realized by stacking semiconductor elements using different semiconductor materials and which has stable and high electrical characteristics, and a method for manufacturing the semiconductor device are provided. be able to.

In addition, by applying the structure of the transistor described in Embodiment 2, a NOR-type circuit and a NAND-type circuit with high reliability and stable characteristics can be provided.

Note that although an example of a NOR circuit and a NAND circuit using the transistor described in Embodiment 2 is described in this embodiment, it is not particularly limited, and an AND circuit, an OR circuit, or the like can also be formed. .

The structures, methods, and the like described in this embodiment can be combined as appropriate with any of the structures, methods, and the like described in the other embodiments.

(Embodiment 5)
In this embodiment, an example of a semiconductor device (storage device) that uses the transistor described in Embodiment 2 and can hold stored data even when power is not supplied and has no limit on the number of writing times. This will be described with reference to the drawings.

FIG. 24A is a circuit diagram illustrating the semiconductor device of this embodiment.

A transistor using a semiconductor material other than an oxide semiconductor (eg, silicon) can be used as the transistor 260 illustrated in FIG. 24A, and high-speed operation is easy. For the transistor 262, a transistor including an oxide semiconductor stack and having a structure similar to that of the transistor described in Embodiment 2 can be used, and the charge can be held for a long time due to the characteristics.

Note that although the above transistors are all assumed to be n-channel transistors, p-channel transistors can be used as transistors used in the semiconductor device described in this embodiment.

In FIG. 24A, the first wiring (1st Line) and the source electrode layer of the transistor 260 are electrically connected, and the second wiring (2nd Line) and the drain electrode layer of the transistor 260 are electrically connected. Connected. Further, the third wiring (3rd Line) and one of the source electrode layer and the drain electrode layer of the transistor 262 are electrically connected, and the fourth wiring (4th Line) and the gate electrode layer of the transistor 262 are connected to each other. Are electrically connected. The other of the gate electrode layer of the transistor 260 and the source or drain electrode layer of the transistor 262 is electrically connected to one of the electrodes of the capacitor 264, and the fifth wiring (5th Line) and the capacitor The other of the H.264 electrodes is electrically connected.

In the semiconductor device illustrated in FIG. 24A, information can be written, held, and read as follows by utilizing the feature that the potential of the gate electrode layer of the transistor 260 can be held.

Information writing and holding will be described. First, the potential of the fourth wiring is set to a potential at which the transistor 262 is turned on, so that the transistor 262 is turned on. Accordingly, the potential of the third wiring is supplied to the gate electrode layer of the transistor 260 and the capacitor 264. That is, predetermined charge is supplied to the gate electrode layer of the transistor 260 (writing). Here, it is assumed that one of two charges (hereinafter, referred to as low level charge and high level charge) that gives two different potential levels is given. After that, the potential of the fourth wiring is set to a potential at which the transistor 262 is turned off and the transistor 262 is turned off, whereby the charge given to the gate electrode layer of the transistor 260 is held (held).

Since the off-state current of the transistor 262 is extremely small, the charge of the gate electrode layer of the transistor 260 is held for a long time.

Next, reading of information will be described. When an appropriate potential (read potential) is applied to the fifth wiring in a state where a predetermined potential (constant potential) is applied to the first wiring, according to the amount of charge held in the gate electrode layer of the transistor 260, The second wiring takes different potentials. In general, when the transistor 260 is an n-channel transistor, the apparent threshold V _{th_H in the} case where a high-level charge is applied to the gate electrode layer of the transistor 260 is a low-level charge applied to the gate electrode layer of the transistor 260. This is because it becomes lower than the apparent threshold value V _{th_L in the} case of Here, the apparent threshold voltage refers to the potential of the fifth wiring necessary for turning on the transistor 260. Therefore, the charge given to the gate electrode layer of the transistor 260 can be determined by _setting the potential of the fifth wiring to a potential V _{0 that} is intermediate between V _{th_H} and V _{th_L} . For example, in the case where a high-level charge is applied in writing, the transistor 260 is turned “on” if the potential of the fifth wiring is V ₀ (> V _{th_H} ). In the case where the low-level charge is _supplied , the transistor 260 remains in the “off state” even when the potential of the fifth wiring is V ₀ (<V _{th_L} ). Therefore, the held information can be read by looking at the potential of the second wiring.

Note that in the case of using memory cells arranged in an array, it is necessary to read only information of a desired memory cell. In the case where information is not read out in this manner, a potential at which the transistor 260 is turned “off” regardless of the state of the gate electrode layer, that is, a potential lower than V _{th_H} may be _supplied to the fifth wiring. Alternatively, a potential at which the transistor 260 is turned on regardless of the state of the gate electrode layer, that is, a potential higher than V _{th_L} may be _supplied to the fifth wiring.

FIG. 24B illustrates an example of one embodiment of a structure of a different memory device. FIG. 24B illustrates an example of a circuit configuration of a semiconductor device, and FIG. 24C is a conceptual diagram illustrating an example of a semiconductor device. First, the semiconductor device illustrated in FIG. 24B is described, and then the semiconductor device illustrated in FIG. 24C is described below.

In the semiconductor device illustrated in FIG. 24B, the bit line BL and the source or drain electrode of the transistor 262 are electrically connected, and the word line WL and the gate electrode layer of the transistor 262 are electrically connected. A source electrode or a drain electrode 262 and the first terminal of the capacitor 254 are electrically connected.

The transistor 262 including an oxide semiconductor has a feature of extremely low off-state current. Therefore, when the transistor 262 is turned off, the potential of the first terminal of the capacitor 254 (or the charge accumulated in the capacitor 254) can be held for an extremely long time.

Next, the case where data is written and held in the semiconductor device (memory cell 250) illustrated in FIG. 24B is described.

First, the potential of the word line WL is set to a potential at which the transistor 262 is turned on, so that the transistor 262 is turned on. Accordingly, the potential of the bit line BL is supplied to the first terminal of the capacitor 254 (writing). After that, the potential of the first terminal of the capacitor 254 is held (held) by setting the potential of the word line WL to a potential at which the transistor 262 is turned off and the transistor 262 being turned off.

Since the off-state current of the transistor 262 is extremely small, the potential of the first terminal of the capacitor 254 (or charge accumulated in the capacitor) can be held for a long time.

Next, reading of information will be described. When the transistor 262 is turned on, the bit line BL in a floating state and the capacitor 254 are brought into conduction, and charge is redistributed between the bit line BL and the capacitor 254. As a result, the potential of the bit line BL changes. The amount of change in the potential of the bit line BL varies depending on the potential of the first terminal of the capacitor 254 (or the charge accumulated in the capacitor 254).

For example, the potential of the first terminal of the capacitor 254 is V, the capacitor of the capacitor 254 is C, the capacitor component of the bit line BL (hereinafter also referred to as bit line capacitor) is CB, and before the charge is redistributed. When the potential of the bit line BL is VB0, the potential of the bit line BL after the charge is redistributed is (CB × VB0 + C × V) / (CB + C). Therefore, when the potential of the first terminal of the capacitor 254 assumes two states of V1 and V0 (V1> V0) as the state of the memory cell 250, the potential of the bit line BL when the potential V1 is held. It can be seen that (= CB × VB0 + C × V1) / (CB + C) is higher than the potential of the bit line BL when the potential V0 is held (= CB × VB0 + C × V0) / (CB + C)).

Then, information can be read by comparing the potential of the bit line BL with a predetermined potential.

As described above, the semiconductor device illustrated in FIG. 24B can hold charge that is accumulated in the capacitor 254 for a long time because the off-state current of the transistor 262 is extremely small. That is, the refresh operation is not necessary or the frequency of the refresh operation can be extremely low, so that power consumption can be sufficiently reduced. Further, stored data can be retained for a long time even when power is not supplied.

Next, the semiconductor device illustrated in FIG. 24C is described.

A semiconductor device illustrated in FIG. 24C includes a memory cell array 251a and a memory cell array 251b each including a plurality of memory cells 250 illustrated in FIG. 24B as memory circuits in an upper portion, and a memory cell array 251 (memory cell array) in a lower portion. 251a and the memory cell array 251b) have a peripheral circuit 253 necessary for operating. Note that the peripheral circuit 253 is electrically connected to the memory cell array 251.

With the structure illustrated in FIG. 24C, the peripheral circuit 253 can be provided immediately below the memory cell array 251 (the memory cell array 251a and the memory cell array 251b), so that the semiconductor device can be downsized.

The transistor provided in the peripheral circuit 253 is preferably formed using a semiconductor material different from that of the transistor 262. For example, silicon, germanium, silicon germanium, silicon carbide, gallium arsenide, or the like can be used, and a single crystal semiconductor is preferably used. In addition, an organic semiconductor material or the like may be used. A transistor using such a semiconductor material can operate at a sufficiently high speed. Therefore, various transistors (logic circuit, drive circuit, etc.) that require high-speed operation can be suitably realized by the transistor.

Note that in the semiconductor device illustrated in FIG. 24C, the structure in which the two memory cell arrays 251 (the memory cell array 251a and the memory cell array 251b) are stacked is illustrated; however, the number of stacked memory cells is not limited thereto. . A configuration in which three or more memory cells are stacked may be employed.

By using a transistor in which an oxide semiconductor is used for a channel formation region as the transistor 262, stored data can be retained for a long time. In other words, since it is possible to obtain a semiconductor memory device that does not require a refresh operation or has a very low frequency of the refresh operation, power consumption can be sufficiently reduced.

Further, by applying the semiconductor device including the oxide stack described in Embodiment 1 as a semiconductor device described in this embodiment and having an oxide semiconductor layer serving as a channel region away from the surface of the oxide stack, A semiconductor device with high reliability and stable electrical characteristics can be obtained.

(Embodiment 6)
In this embodiment, an example in which the semiconductor device described in any of the above embodiments is applied to an electronic device such as a mobile phone, a smartphone, or an e-book reader will be described with reference to FIGS.

FIG. 25 is a block diagram of an electronic device. 25 includes an RF circuit 901, an analog baseband circuit 902, a digital baseband circuit 903, a battery 904, a power supply circuit 905, an application processor 906, a flash memory 910, a display controller 911, a memory circuit 912, a display 913, and a touch. A sensor 919, an audio circuit 917, a keyboard 918, and the like are included. The display 913 includes a display unit 914, a source driver 915, and a gate driver 916. The application processor 906 includes a CPU 907, a DSP 908, and an interface (IF) 909. In general, the memory circuit 912 includes an SRAM or a DRAM, and by using the semiconductor device described in the above embodiment for this portion, information can be written and read at high speed, and long-term storage can be performed. In addition, a highly reliable electronic device in which power consumption is sufficiently reduced can be provided.

FIG. 26 shows an example in which the semiconductor device described in any of the above embodiments is used for the memory circuit 950 of the display. A memory circuit 950 illustrated in FIG. 26 includes a memory 952, a memory 953, a switch 954, a switch 955, and a memory controller 951. The memory circuit reads a signal line from the image data (input image data), data (stored image data) stored in the memory 952 and the memory 953, and a display controller 956 that performs control, and a display controller 956 A display 957 for displaying by signals is connected.

First, certain image data is formed by an application processor (not shown) (input image data A). The input image data A is stored in the memory 952 via the switch 954. The image data (stored image data A) stored in the memory 952 is sent to the display 957 via the switch 955 and the display controller 956 and displayed.

When there is no change in the input image data A, the stored image data A is normally read from the display controller 956 from the memory 952 via the switch 955 at a cycle of about 30 to 60 Hz.

Next, for example, when the user performs an operation of rewriting the screen (that is, when the input image data A is changed), the application processor forms new image data (input image data B). The input image data B is stored in the memory 953 via the switch 954. During this time, stored image data A is periodically read from the memory 952 via the switch 955. When new image data (stored image data B) is stored in the memory 953, the stored image data B is read from the next frame of the display 957, and is stored in the display 957 via the switch 955 and the display controller 956. The stored image data B is sent and displayed. This reading is continued until new image data is stored in the memory 952 next time.

As described above, the memory 952 and the memory 953 display the display 957 by alternately writing image data and reading image data. Note that the memory 952 and the memory 953 are not limited to different memories, and one memory may be divided and used. By employing the semiconductor device described in any of the above embodiments for the memory 952 and the memory 953, information can be written and read at high speed, data can be stored for a long time, and power consumption can be sufficiently reduced. it can. In addition, a highly reliable semiconductor device which is not easily affected by intrusion of water, moisture, or the like from the outside can be provided.

FIG. 27 shows a block diagram of an electronic book. 27 includes a battery 1001, a power supply circuit 1002, a microprocessor 1003, a flash memory 1004, an audio circuit 1005, a keyboard 1006, a memory circuit 1007, a touch panel 1008, a display 1009, and a display controller 1010.

Here, the semiconductor device described in any of the above embodiments can be used for the memory circuit 1007 in FIG. The memory circuit 1007 has a function of temporarily holding the contents of a book. For example, when the user uses the highlight function, the memory circuit 1007 stores and holds information on a location designated by the user. The highlight function is marking by marking a specific part when the user is reading an e-book, for example, changing the display color, underlining, thickening the character, or changing the typeface of the character. To show the difference from the surroundings. The memory circuit 1007 may be used for short-term information storage, and data stored in the memory circuit 1007 may be copied to the flash memory 1004 for long-term information storage. Even in such a case, by adopting the semiconductor device described in the above embodiment, writing and reading of information can be performed at high speed, long-term memory retention can be performed, and power consumption can be sufficiently reduced. Can do. In addition, a highly reliable semiconductor device which is not easily affected by intrusion of water, moisture, or the like from the outside can be provided.

FIG. 28 shows a specific example of an electronic device. 28A and 28B illustrate a tablet terminal that can be folded. In FIG. FIG. 28A illustrates an open state in which a tablet terminal includes a housing 9630, a display portion 9631a, a display portion 9631b, a display mode switching switch 9034, a power switch 9035, a power saving mode switching switch 9036, and a fastener 9033. And an operation switch 9038.

The semiconductor device described in the above embodiment can be used for the display portion 9631a and the display portion 9631b, so that a highly reliable tablet terminal can be obtained. Further, the memory device described in the above embodiment may be applied to the semiconductor device in this embodiment.

Part of the display portion 9631 a can be a touch panel region 9632 a and data can be input when a displayed operation key 9638 is touched. Note that in the display portion 9631a, a structure in which half of the regions have a display-only function and a structure in which the other half has a touch panel function is shown as an example; however, the structure is not limited thereto. The entire surface of the display portion 9631a can display keyboard buttons to serve as a touch panel, and the display portion 9631b can be used as a display screen.

Further, in the display portion 9631b, as in the display portion 9631a, part of the display portion 9631b can be a touch panel region 9632b. Further, a keyboard button can be displayed on the display portion 9631b by touching a position where the keyboard display switching button 9539 on the touch panel is displayed with a finger or a stylus.

Touch input can be performed simultaneously on the touch panel region 9632a and the touch panel region 9632b.

A display mode switching switch 9034 can switch the display direction such as vertical display or horizontal display, and can select switching between monochrome display and color display. The power saving mode change-over switch 9036 can optimize the display luminance in accordance with the amount of external light during use detected by an optical sensor built in the tablet terminal. The tablet terminal may include not only an optical sensor but also other detection devices such as a gyroscope, an acceleration sensor, and other sensors that detect inclination.

FIG. 28A illustrates an example in which the display areas of the display portion 9631b and the display portion 9631a are the same, but there is no particular limitation, and one size may differ from the other size, and the display quality may also be different. May be different. For example, one display panel may be capable of displaying images with higher definition than the other.

FIG. 28B illustrates a closed state, in which the tablet terminal includes a housing 9630, a solar battery 9633, a charge / discharge control circuit 9634, a battery 9635, and a DCDC converter 9636. Note that FIG. 12B illustrates a structure including a battery 9635 and a DCDC converter 9636 as an example of the charge / discharge control circuit 9634.

Note that since the tablet terminal can be folded in two, the housing 9630 can be closed when not in use. Accordingly, since the display portion 9631a and the display portion 9631b can be protected, a tablet terminal with excellent durability and high reliability can be provided from the viewpoint of long-term use.

In addition, the tablet terminal shown in FIGS. 28A and 28B has a function for displaying various information (still images, moving images, text images, etc.), a calendar, date or time, and the like. A function for displaying on the display unit, a touch input function for performing touch input operation or editing of information displayed on the display unit, a function for controlling processing by various software (programs), and the like can be provided.

The structures, methods, and the like described in this embodiment can be combined as appropriate with any of the structures, methods, and the like described in the other embodiments.

In this example, a result obtained by forming a conductive film over an oxide semiconductor film and examining element movement between stacked films by SIMS analysis will be described.

FIGS. 11A and 11B show the results of the SIMS analysis of the oxygen isotope ( ¹⁸ O) profile in the depth direction before and after the heat treatment by fabricating a stacked sample of an IGZO film and a tungsten film using a sputtering method. is there. Note that the IGZO film uses In: Ga: Zn = 1: 1: 1 or 1: 3: 2 (atomic ratio) as a sputtering target, and Ar: O ₂ ( ¹⁸ O) = 2: 1 (flow ratio). It is formed by a DC sputtering method using it as a film forming gas. The tungsten film was formed by a DC sputtering method using metallic tungsten as a sputtering target and Ar 100% as a deposition gas. The heat treatment was performed at 300 ° C., 350 ° C., 400 ° C., and 450 ° C. for 1 hour each, and comparison was performed on each of five samples including the sample that was not subjected to heat treatment.

Here, the IGZO film formed using In: Ga: Zn = 1: 1: 1 (atomic ratio) as a sputtering target is a CAAC-OS film, and In: Ga: Zn = 1: 3: 2 (number of atoms). IGZO film formed as a sputtering target is an amorphous IGZO film.

As shown in FIGS. 11A and 11B, regardless of the composition (crystallinity) of the oxide semiconductor film, oxygen in the oxide semiconductor film may be taken into the tungsten film side when the heat treatment temperature is increased. Recognize.

Since there are several heating steps in the manufacturing process of the transistor, oxygen vacancies are generated in a region in contact with the source electrode and the drain electrode of the oxide semiconductor layer due to the above phenomenon, and the region becomes n-type. . Therefore, the n-type region can function as a source or a drain of the transistor.

12A and 12B show the results of SIMS analysis of a sample manufactured using a tantalum nitride film instead of the tungsten film. The tantalum nitride film was formed by a reactive sputtering method (DC sputtering method) using metal tantalum as a sputtering target and Ar: N ₂ = 5: 1 (flow rate ratio) as a film forming gas. In addition, as heat processing, it carried out on each of 4 conditions similar to the above, and compared with each of 5 samples including the sample which has not performed heat processing.

FIG. 12A shows SIMS analysis results of a stacked sample of an IGZO film and a tantalum nitride film of In: Ga: Zn = 1: 1: 1. None of the samples showed any movement (incorporation) of oxygen into the tantalum nitride film, and showed a behavior different from that of the tungsten film shown in FIG. FIG. 12B shows SIMS analysis results of a stacked sample of an IGZO film and a tantalum nitride film of In: Ga: Zn = 1: 3: 2. None of the samples showed any movement (incorporation) of oxygen into the tantalum nitride film, and showed a behavior different from that of the tungsten film shown in FIG. Therefore, it can be said that the tantalum nitride film is a film that does not easily bond to oxygen or a film that hardly absorbs oxygen.

FIGS. 13A and 13B show the results of SIMS analysis of a sample manufactured using a titanium nitride film instead of the tungsten film. The titanium nitride film was formed by a reactive sputtering method (DC sputtering method) using metal titanium as a sputtering target and N ₂ 100% as a film forming gas. The heat treatment was performed under the same four conditions as described above, and a comparison was made with each of the five samples including the sample not subjected to the heat treatment.

FIG. 13A shows SIMS analysis results of a stacked sample of an IGZO film and a titanium nitride film with In: Ga: Zn = 1: 1: 1. None of the samples showed any movement (incorporation) of oxygen into the titanium nitride film, and showed a behavior different from that of the tungsten film shown in FIG. FIG. 13B shows SIMS analysis results of a stacked sample of an IGZO film and a titanium nitride film with In: Ga: Zn = 1: 3: 2. None of the samples showed the movement (incorporation) of oxygen into the titanium nitride film, and showed a behavior different from that of the tungsten film shown in FIG. Therefore, it can be said that the titanium nitride film is a film that hardly binds to oxygen or a film that hardly takes up oxygen.

Next, the results of investigation by SIMS analysis on the migration of impurities into the IGZO film will be described.

14A and 14B show the results of SIMS analysis of the profile in the depth direction of nitrogen before and after the heat treatment, in which a tantalum nitride or titanium nitride film is formed on the IGZO film by a sputtering method. Note that the IGZO film is formed by DC sputtering using In: Ga: Zn = 1: 1: 1 (atomic ratio) as a sputtering target and Ar: O ₂ = 2: 1 (flow ratio) as a deposition gas. Formed. Further, the tantalum nitride film and the titanium nitride film were formed by the above-described manufacturing method. In addition, heat processing was performed on 400 degreeC and the conditions for 1 hour, and each 2 samples including the sample which has not performed heat processing were compared.

As shown in FIGS. 14A and 14B, it was found that no migration of nitrogen into the IGZO film was confirmed in any sample. Therefore, it was found that nitrogen serving as a donor in the IGZO film does not move widely from the tantalum nitride and titanium nitride films into the IGZO film, and thus does not make the channel formation region of the transistor n-type.

FIGS. 15A and 15B show the results of SIMS analysis of the profile in the depth direction of Ta or Ti for the same sample illustrated in FIG. As shown in FIGS. 15A and 15B, it was found that Ta or Ti did not move into the IGZO film. Therefore, it has been found that Ti and Ta, which can be impurities affecting the electric characteristics of the transistor, do not move widely from the tantalum nitride film or the titanium nitride film into the IGZO film.

As described above, a conductive nitride such as tantalum nitride or titanium nitride is a film that does not easily bond to oxygen or a film that does not easily absorb oxygen, and nitrogen and a metal element in the conductive nitride film are contained in the oxide semiconductor film. It was shown that it was difficult to take in.

This example can be implemented in combination with any of the other embodiments described in this specification as appropriate.

In this example, a result obtained by forming a conductive film over an oxide semiconductor film, removing the conductive film, and measuring the sheet resistance value of the oxide semiconductor film will be described.

FIG. 16 shows a sample manufactured by forming an IGZO film using a sputtering method, stacking the IGZO film on the IGZO film, forming a tungsten film or a titanium nitride film by a sputtering method, and then removing the tungsten film or the titanium nitride film. It is the result of having measured the sheet resistance value with respect to the depth which etched the IGZO film | membrane. For comparison, a sample in which a conductive film was not formed over the IGZO film was also produced. Note that the IGZO film uses In: Ga: Zn = 1: 1: 1 (atomic ratio) as a sputtering target and Ar: O ₂ ( ¹⁸ O) = 2: 1 (flow ratio) as a deposition gas. It formed by DC sputtering method. The tungsten film was formed by a DC sputtering method using metallic tungsten as a sputtering target and Ar 100% as a deposition gas. The titanium nitride film was formed by a reactive sputtering method (DC sputtering method) using metal titanium as a sputtering target and N ₂ 100% as a film forming gas. Hydrogen peroxide water was used for etching the tungsten film and the titanium nitride film. For the etching of the IGZO film, a mixed aqueous solution of hydrogen peroxide and ammonia was used. The etching depth of the IGZO film was determined from the thickness of the remaining film measured using spectroscopic ellipsometry before and after etching.

As shown in FIG. 16, in the sample in which the tungsten film was formed on the IGZO film, it was confirmed that the resistance was reduced to a depth of about 5 nm from the surface of the IGZO film. This is because a low-resistance mixed layer of IGZO and tungsten is formed near the surface of the IGZO film, or oxygen in the IGZO film is taken into the tungsten film, resulting in oxygen deficiency near the surface of the IGZO film. This suggests that an n-type region is formed.

On the other hand, in the sample in which titanium nitride was formed on the IGZO film and the sample in which the conductive film was not formed, the resistance reduction of the IGZO film could not be confirmed. This suggests that elements constituting titanium nitride are not easily taken into the IGZO film, or that oxygen in the IGZO film is hard to be taken into the titanium nitride film.

FIG. 17A illustrates an example in which an IGZO film is formed using a sputtering method, stacked on the IGZO film to form a tungsten film or a titanium nitride film, and then subjected to heat treatment, and then the tungsten film or the titanium nitride film is subjected to heat treatment. It is the result of having measured the sheet resistance value with respect to the depth which etched the IGZO film | membrane about the sample produced by removing a film | membrane. For comparison, a sample in which a conductive film was not formed over the IGZO film was also produced. Note that formation and removal of the IGZO film, the tungsten film, and the titanium nitride film were performed in the same manner as described above. The heat treatment was performed under a condition of 400 ° C. for 1 h in an N ₂ atmosphere.

As shown in FIG. 17A, it was confirmed that the resistance of the IGZO film was lowered in any of the samples. Here, it was confirmed that the sample in which the tungsten film was formed on the IGZO film had the lowest resistance near the surface and the lowest resistance. This indicates that the tungsten film is most likely to take up oxygen in the IGZO film. In addition, the sample in which titanium nitride is formed on the IGZO film shows the same behavior as the sample in which the conductive film is not formed on the IGZO film. That is, in the sample in which the tungsten film is formed on the IGZO film, the resistance of the IGZO film is lowered by the oxygen in the IGZO film being taken into the tungsten film, whereas the titanium nitride film is formed on the IGZO film. The sample suggests that oxygen released from the IGZO film passes through the titanium nitride film and is released upward. This result is in good agreement with the SIMS analysis result shown in Example 1.

In FIG. 17B, a silicon oxide film is formed by a sputtering method, an IGZO film is formed on the silicon oxide film by a sputtering method, and a tungsten film or a titanium nitride film is formed by stacking the IGZO film by a sputtering method. Then, the sheet resistance value with respect to the depth obtained by etching the IGZO film is measured for a sample manufactured by removing the tungsten film or the titanium nitride film after the heat treatment. For comparison, a sample in which a conductive film was not formed over the IGZO film was also produced. The silicon oxide film was formed by a reactive sputtering method (DC sputtering method) using silicon as a sputtering target and O ₂ 100% as a film forming gas. Note that formation and removal of the IGZO film, the tungsten film, and the titanium nitride film were performed in the same manner as described above. The heat treatment was performed under a condition of 400 ° C. for 1 h in an N ₂ atmosphere.

In FIG. 17B, it can be confirmed that the region in which the resistance of the IGZO film is reduced is shallower in the thickness direction than the result shown in FIG. This indicates that oxygen is supplied from the silicon oxide film to the IGZO film by the heat treatment, thereby reducing the oxygen vacancies in the IGZO film and increasing the resistance of the IGZO film. Thus, it was found that the thickness of the region of the IGZO film where the resistance is reduced can be controlled by using a film capable of releasing oxygen below the IGZO film.

As described above, it has been confirmed that by forming a conductive film that easily takes in oxygen, such as a tungsten film, in contact with the IGZO film, a region in the vicinity of the IGZO film in contact with the conductive film can be reduced in resistance. Furthermore, it has been confirmed that by performing the heat treatment, the region of the IGZO film where the resistance is reduced can be expanded in the depth direction. It was also found that the thickness of the region where the resistance is reduced can be controlled by forming a film capable of releasing oxygen in the vicinity of the IGZO film.

This example can be implemented in combination with any of the other embodiments described in this specification as appropriate.

In this example, a sample in which oxygen is added to an oxide semiconductor film by an ion implantation method, the result of TDS (Thermal Desorption Spectroscopy) analysis, and evaluation of the film density will be described.

First, the produced sample will be described. A silicon oxynitride film was formed by a plasma CVD method on a silicon wafer subjected to thermal oxidation treatment in an HCl atmosphere, and then the surface of the silicon oxynitride film was planarized by a CMP method. After that, an IGZO film was formed on the silicon oxynitride film, and oxygen ions (O ⁺ ) were added to the IGZO film by an ion implantation method. Here, the IGZO film is formed by DC sputtering using In: Ga: Zn = 1: 3: 2 (atomic ratio) as a sputtering target and Ar: O ₂ = 2: 1 (flow ratio) as a film forming gas. Formed. Oxygen ions were performed under the conditions of an acceleration voltage of 5 kV and a dose of 1.0 × 10 ¹⁶ ions / cm ² . For comparison, a sample to which oxygen ions were not added was also prepared.

FIG. 18A shows a TDS analysis result obtained by measuring a release amount of a gas having a mass number of 32 in a sample to which oxygen ions were not added. In the range from about 50 ° C. to about 550 ° C., no release peak of a gas having a mass number of 32 was confirmed. FIG. 18B shows a TDS analysis result obtained by measuring a release amount of a gas having a mass number of 32 in a sample to which oxygen ions were added. A remarkable release peak was observed at about 400 ° C. to about 500 ° C. Accordingly, by adding oxygen ions to the oxide semiconductor film, excess oxygen can be retained in the oxide semiconductor film, and further, by heating the oxide semiconductor film to which oxygen ions are added. It has been found that excess oxygen is released from the oxide semiconductor film. Therefore, by providing such an oxide semiconductor film to which oxygen is added in contact with the oxide semiconductor layer which forms a channel of the transistor, by performing heat treatment on the oxide semiconductor film, an oxide semiconductor to which oxygen is added is provided. Oxygen can be effectively supplied from the film to the oxide semiconductor layer forming the channel.

Further, the oxide semiconductor film density was evaluated using an X-ray reflectivity measurement method (XRR: X-Ray Reflectivity) for a sample to which oxygen ions were not added and a sample to which oxygen ions were added. .

First, the produced sample will be described. A silicon oxynitride film is formed by plasma CVD on a silicon wafer that has been subjected to thermal oxidation in an HCl atmosphere, oxygen ions (O ⁺ ) are added to the silicon oxynitride film by ion implantation, and then The surface of the silicon oxynitride film was planarized by a CMP method. After that, an IGZO film was formed on the silicon oxynitride film, and oxygen ions (O ⁺ ) were added to the IGZO film by an ion implantation method. Here, formation of the IGZO film and addition of oxygen ions to the IGZO film were performed under the same conditions as described above. Addition of oxygen ions to the silicon oxynitride film was performed under the conditions of an acceleration voltage of 60 kV and a dose of 2.0 × 10 ¹⁶ ions / cm ² . For comparison, a sample to which oxygen ions were not added was also prepared.

As a result of measuring the film density by XRR, the sample without oxygen ion addition has a film density value of 5.8 g / cm ³ , and the sample with oxygen ion addition has a film density value of 5 It was 0.6 g / cm ³ , and it was found that the film density of the oxide semiconductor film was reduced by the addition of oxygen ions. This suggests that the addition of oxygen ions makes the oxide semiconductor film more disordered atomic arrangement, that is, it can be modified into a remarkable amorphous film. For example, in the case where an oxide semiconductor film having crystallinity is stacked over an oxide semiconductor film, when a crystalline oxide semiconductor film having a different crystal structure is used as a lower layer, the oxide semiconductor film formed in the upper layer Crystallinity may decrease. Even in such a case, the oxide semiconductor film with improved crystallinity can be obtained by adding oxygen ions to the lower oxide semiconductor film to make it amorphous before forming the upper oxide semiconductor film. Can be formed in the upper layer.

This example can be implemented in combination with any of the other embodiments described in this specification as appropriate.

140 resist mask 200 region 302 oxygen 310 transistor 320 transistor 330 transistor 340 transistor 400 substrate 402 base insulating layer 404 oxide stack 404a first oxide layer 404b oxide semiconductor layer 404c second oxide layer 404d third oxide Layer 405 region 406a first source electrode layer 406b first drain electrode layer 408a second source electrode layer 408b second drain electrode layer 410 gate insulating layer 412 gate electrode layer 412a first gate electrode layer 412b second Gate electrode layer 414 Protective insulating layer 436a Oxide film 436b Oxide semiconductor film 436c Oxide film 436d Oxide layer 437d Oxide layer 438d Oxide layer 456a Oxide layer 456b Oxide semiconductor layer 456c Oxide layer 456d Oxide layer

Claims (6)

A base insulating layer containing oxygen;
An island-shaped oxide stack provided on the base insulating layer;
A first source electrode layer and a first drain electrode layer in contact with a part of the upper surface of the island-shaped oxide stack and a side surface in the channel formation direction;
A second source electrode layer and a second drain electrode made of a metal nitride film are provided on the first source electrode layer and the first drain electrode layer, respectively, and are in contact with part of the upper surface of the oxide stack. Layers,
A gate insulating layer provided on the second source electrode layer and the second drain electrode layer and in contact with the upper surface of the oxide stack between the second source electrode layer and the second drain electrode layer; When,
A gate electrode layer overlapping with the oxide stack via the gate insulating layer;
A protective insulating layer provided on and in contact with the gate insulating layer and the gate electrode layer,
The oxide stack includes at least an oxide semiconductor layer that forms a channel, a first oxide layer provided between the oxide semiconductor layer and the base insulating layer, the oxide semiconductor layer, and the gate. A second oxide layer provided between the insulating layer and a third oxide provided in contact with a side surface of the first oxide layer, the oxide semiconductor layer, and the second oxide layer; And the third oxide layer has a curved surface,
The base insulating layer and the gate insulating layer have a region in contact with an outer peripheral portion of the island-shaped oxide stack,
The protective insulating layer is a semiconductor device that is less permeable to oxygen than the gate insulating layer. A base insulating layer containing oxygen;
An island-shaped oxide stack provided on the base insulating layer;
A first source electrode layer and a first drain electrode layer in contact with a part of the upper surface of the island-shaped oxide stack and a side surface in the channel formation direction;
A second source electrode layer and a second drain electrode made of a metal nitride film are provided on the first source electrode layer and the first drain electrode layer, respectively, and are in contact with part of the upper surface of the oxide stack. Layers,
A gate insulating layer provided on the second source electrode layer and the second drain electrode layer and in contact with an upper surface of the oxide stack between the second source electrode layer and the second drain electrode layer; ,
A gate electrode layer overlapping with part of the oxide stack, the second source electrode layer, and the second drain electrode layer through the gate insulating layer;
The oxide stack includes at least an oxide semiconductor layer that forms a channel, a first oxide layer provided between the oxide semiconductor layer and the base insulating layer, the oxide semiconductor layer, and the gate. A second oxide layer provided between the insulating layer and a third oxide provided in contact with a side surface of the first oxide layer, the oxide semiconductor layer, and the second oxide layer; And the third oxide layer has a curved surface,
The base insulating layer and the gate insulating layer have a region in contact with an outer peripheral portion of the island-shaped oxide stack,
The protective insulating layer is a semiconductor device that is less permeable to oxygen than the gate insulating layer. In claim 1 or claim 2,
The oxide semiconductor layer, the first oxide layer, and the second oxide layer include an In-M-Zn oxide (M is Al, Ti, Ga, Y, Zr, La, Ce, Nd, or Hf). ) And
The first oxide layer and the second oxide layer are semiconductor devices in which an atomic ratio of M to In is larger than that of the oxide semiconductor layer. In any one of Claim 1 thru | or 3,
The oxide semiconductor layer includes a crystal part,
A semiconductor device in which a c-axis of the crystal part is parallel to a normal vector of a surface of the oxide semiconductor layer. In any one of Claims 1 thru | or 4,
A semiconductor device using a material that is more easily bonded to oxygen than the second source electrode layer and the second drain electrode layer for the first source electrode layer and the first drain electrode layer. In any one of Claims 1 thru | or 5,
A semiconductor device in which a hydrogen concentration contained in the protective insulating layer is less than 5 × 10 ¹⁹ cm ⁻³ .